From f1785b7b656a162e4a4c90830e3a8df87b54b484 Mon Sep 17 00:00:00 2001 From: "Chanwut (Mick) Kittivorawong" Date: Sun, 6 Mar 2022 20:16:46 -0800 Subject: [PATCH 1/4] squash graph --- .gitignore | 6 +- amber_videos/traffic-scene-mini.mp4 | Bin 0 -> 165978 bytes apperception/apperception_benchmark.py | 9 +- apperception/lens.py | 17 + apperception/new_apperception_benchmark.py | 135 ++++ apperception/new_compute_lib.py | 18 + apperception/new_db.py | 381 +++++++++++ apperception/new_readme.md | 24 + apperception/new_util.py | 194 ++++++ apperception/new_world.py | 627 ++++++++++++++++++ .../object_tracker_yolov5_deepsort.py | 48 +- apperception/point.py | 29 +- apperception/scenic_benchmark.py | 86 +++ apperception/test_parse_scenic_pred.py | 95 +++ apperception/video_context.py | 1 + apperception/video_util.py | 512 +++++++++++++- apperception/world.py | 1 + apperception_example.ipynb | 62 +- new_apperception_benchmark.ipynb | 426 ++++++++++++ scenic_equivelants.ipynb | 588 ++++++++++++++++ scenic_equivelants.py | 367 ++++++++++ setup.sh | 6 +- 22 files changed, 3565 insertions(+), 67 deletions(-) create mode 100644 amber_videos/traffic-scene-mini.mp4 create mode 100644 apperception/new_apperception_benchmark.py create mode 100644 apperception/new_compute_lib.py create mode 100644 apperception/new_db.py create mode 100644 apperception/new_readme.md create mode 100644 apperception/new_util.py create mode 100644 apperception/new_world.py create mode 100644 apperception/scenic_benchmark.py create mode 100644 apperception/test_parse_scenic_pred.py create mode 100644 new_apperception_benchmark.ipynb create mode 100644 scenic_equivelants.ipynb create mode 100644 scenic_equivelants.py diff --git a/.gitignore b/.gitignore index d93792c2..6ba389ad 100644 --- a/.gitignore +++ b/.gitignore @@ -1,8 +1,12 @@ __MACOSX __pycache__ +.DS_Store +.ipynb_checkpoints yolov4-deepsort yolov5-deepsort .mypy_cache .idea -.ipynb_checkpoints +output +.apperception_cache +env diff --git a/amber_videos/traffic-scene-mini.mp4 b/amber_videos/traffic-scene-mini.mp4 new file mode 100644 index 0000000000000000000000000000000000000000..cb985f702de9e9c39dc798c59dfcb0fec6ce27b6 GIT binary patch literal 165978 zcmeFX^K)ih&^8*|wr$%sCzx1wY)ow1p4hf+TNB&1lauFt&v*WZQ}z9@dpEA`)xEl_ zcI{OQ2nYzz)Y;wM#-5D@2nYz+!pY9&hiCpjvW};KQh>Q)542}7ji5vl@Khy+ZWNl|`#m~&c#KS~nXlrQg z?gZdxa%1KBd5CNP{3aHLL{2V7KO8fWy_5U@Qhi4ger5)yAD16OWMkn5Fwy@n-VYDZ zcQmv$`$2OM8Jj!W*%<2okE^pIz}niv=?B5(#${sc{1eC6!RALa0T`Ni*x3U3S(xdV znTbpdot*XUovbYE|3mz500(=0J5y69fb$RB+1&BR!HJ)Rg~-~@&dSjIN7Da4N>(B# zYYXF_T>g)OiOAORf7URzurYN0uLKKQXMm%%;ScCXZ)EM_Xy~qQY-eL{_+N12pAB+$ zG_S$;KaN=hrGSatq|KTl6_*s5BPtNgDhD}eh?YlKegR?lKnhw1N%ig~c5G!l7({5pD-PJO6IX%L;`)+;upSgO znx(7o1@sV$=R7ox1t~LB8B9v8&>|IA(mryXH*W53TqW2KxwC-JySpq6B>PfZs%J)} zV@96-PVTOq;m$rUVRsOt4QBbk`)50JGpE;_@2F@|&%rZt3?0jDYo zpzE=zEXfNi6W$|RDI6}nelHa=yI0aw%109EDA~jWLlBq_wc4{Q6|lCYsz&9SHT92h z)s?n2a-Ot8hf`42y$Mpo@PVM9> zdqaNttV3J(ZRUW2g1piiS4J+i3IS^8H8vX8L&(WxZ(uQz%&EV(2-yVLR6nF^;~mMh zcJiaPSJQP*45Ri=Io&l4hA!j_K&?PnC+%RQVE!Tp^I2GbPToy#O>yb_aBp{X^(R@G zi85+_7*S?SeS!J|F`r~zDjFZ>BGbHda9jOlm_3MLQMHAU)}$$CbOzcRIv5has_rE(G&o2h}9g5&e`;uS6z7!Vo zQE`j0k6WUyu`ogPJm$E@47>l&UyVaD>!*_3uH#9yUJmPBneN>pHPh;YE z=~}E}XbbDRf&IMN8_x5$br`ZUfV_L)9l_pV*lR$WjdS)OkW`|b8n}{rs3oP-iP0V`DWdocpir!&v8-Jmm{$O2q8PD?XtiRfWTOy&67FcJ5?7WEos znEr@p{G7e!^oI5~8?T#Y{F2sC-<79GF!5s+cvXAiNTLCvw4MovI(6y@c=!5Kun$*Y z&QioT!t96RA4s<2cnt-?U&qWe`u}=$UQ~p8<5NuSKAcw_n&jAG#wlY1Pf7bTjB@fb zKFI9KSIlGD1-aP&{%#CNv2Uh|R;%+zsF;UKPOhH%yi{+35-lm71IHPTCm-~yh=lV~ zLpZ(1Th}rtFND_mFy}7EzHWplP%GVh40LkX^a$?vJ3?4HgWWFR1bXMDAmKwY3EGn# zK%fXu*snR{I9^uZ83$489V-Snrx*Oj->wZNn3AVw$|$b}yIME6HEZ|LUbsRuyY8cF zPfD4a9nGCgw1}z)(XV zES+*X`NR})0@e+}W|0~=oR$U~3tM^?^yx5gol=A8Q;N&feR_!;5n$ud?X|HaL+PuM z9N@PAbRsDLZJ`xwhgs=Q+&NRnd{BsnDb+*@ zG{N26S>1g#Iz^rQgC3+;PrJ?r2o;d0iL@@ypO6LUnCD)uN_xZG^3kfaAVYO)Xx337ekDIBC#cUuPF>RNx1){wwXWd_#qM21O`|ZA)r_W66rG5dglsvFg z#TF|AX^jBU3hx*NiV}omiZ!lOe>tvIE19@n2C1>%hDA5%@l?JyKR@rz(gkAWg6SC~QRrGi+;zKFK%CT5O8dfyxl` z_Qn;LYWBa+J{zs8R|oAi$Rh1KLppMQ9NY8+oy*XX6tY!8fq0phT#qE{B|TVQbQW0f zZPGZxLqXUEr8Py?PWGJgmG!}X302{^ zq!@t*@wZOOuln^}rgd7rT{HaThl6TuNgnim)|>UU{JN9DgtQA( z3Ld0ehDL377vTJlwI*hiM78i6h7uoha&sWQHyU)Y6+!%5Ev<%$yY`3%6F9!B;EsbV zPaaSnG}g@AalHR~o~k>w5~6ks!vP~l;~Np1K}!P^!gIX^)J|pQKOm0O^rpt@cHkhAzgFU={xJmrbhWD zFB$JAqX@QVOP^shaeBwIWfzq>pEz`7r^egiHizB`OZ9c6; z8tqaEIzs$y`vKOUQShlMpYDtrkmIVU4VH#{)^3f$xQ%@sFj=$9PRo9}rr~AZzV)`u{&8Gf1P1Hv!Uju$awG>EzDY<6SITD5nNKf#*W z?q+CW{Q3>0`MR;Kh&A<|)RTTppf%%7AhtP(nFfHvE@r897N|Aa^ zP@uPDZ8%##AA3j`B!7LtIPM@3En~WVqTj`HlncuIP?-e41>**C2n}&zwQpQiDS?(1 z?tLv$vvQSy7--lYtYi(-iMX~%tp5~xz#zwoOWM&B($zk$WMDgna;Oq!n5GFBHTShE z%XG^ITT>6L>{>l{oJWhq1<@woKiehj{=S~LEc70d94wD}Z~}WW4w1Pu+bF<|Xyg_L z`!n6{mDAy5byI@1u(F@n=hzTg&}_FT+n<$WS)&gVv*jlyPUl3Th-5; zBrgDmfJKR(P!^uorO){iVCudzas*Xn;Xdfylcb8-9e2mi-1qjrTP&a2iuV*JR_E;t zi~hm{_C(>T>5-Ad z-^>D}rK&w>Rs93`-vo&U(G}2nTY}IA2dHxvRqHCN5{ToG>`I?$NL@da|-Ys(qWTSby{B<9EYW9 za3pADRJ3r~xR*3j2jej$K>`}DjN4FV!4>CFiVi(P0cbPOrg+S{y}Pzl5O=PJ;~su$ zf_cg(PCAyaZ6 zBTV4`kUNu7=a&z_Qmd)?(7~YWWz6KEt_g1-T)@BQV1^QxmU{W8wNYRNdzdMpcU@&b z29kpWvXIHa*}p`phsd%DY4zEXyE?Y_HCADKXY7o_NvVd8kTidmPOPJhviw;z$a^`z zT%0pHrPN0yOXU{kvnua~a&U2k(A{R!*D$XCQbOJ@C*JWFAONngltv9;lK_OpA_(<$<*gpydGs^*JOp5 zyoO>!gu%v>4i1+XQTe^F&t;So%zv2_Qnk3GTw`ci2zq|`F6z(#Y$%vI!NF!Snp}0v z06E+2+$Ouna()K+CT&QPUP`u;AS*H0IOwP!NwAk9w0YfM;7cg>K#egXpJu#_OxJG+ z(oC?}zc?dyr);ins`%MKWGPSnT^x+`-J;H`LL!7sEjp4PUTc6@)*qk1YJ!B`fy7rJ_;-n+~D#{1+*Aa$N(n2|E zW)$=NLdQAhB?Uuh6AcCN$~?FxrGEZ4^B&cEw-5Sb&=7OxCa#jzvlbuxVAY3&A4+jF z+V%VcNFEe%>!j96Ncn>oV_Ge++)Sl~FE~%qOVn36oK8V! zH+XTGysRBcc05bnXCmFhw7W^sK``|qlm zOq~nt)vLN#XzB0N)O#6!3NX({%u=oryWh&yjStO}=u9MNckOw;K#{0J^cug19>lX{ zK=lN=F?!>}efR3tLav7wtxXZG+hBkp=)WSCcE8;uJKU$&OOiRw`zfB}oK?9VQ4q%z zKPnX$>nBM3z_uj+LJBM z{Ayi{#>V+Q3MA*a3VcmWXdM$4-cFyKD=om^xx0t8`m;Rle0wKlu;k`<9md5Y53;7S z5GS1O@(6ZSPMEHQ>B!OPw4N&b8NaU)&eQr3qkr~_pRDi0njgLhzBJLf^$d7Mc3@ZQ49h>v_+Hm|>LjbG68 zX*i=w*+JUr$^>6uDx~z0iO5T?yr6Djk%(L=+M16tS&PbB-FnVL!9wSk=1sahQYrA2 z=7dz|L=mH~xyzl@?#`Xb@S(mpLi>F{f~;H4nn@$?_lfwFwEGB32fzF&VExW{X!<0t z&p*Gz-&O`>FTA*97{e=+I28=&1;M%L8PkoGTUv%`8^oO=Z@MdStR;a-y1AZ=8@_heBvX&Tu~gTpU<660YFYBRY9nN$?0O^tk{aL=aZLIod)s!qhYf*7NIsN$BJU zp~sCT{$kIEQ*W->8lC!tzPSk22YsQEY@IXbSmifF9}va%zdC)~67|CW=dJMR1`Lx{ zvVgvQZO8NK{t&@m=alm&tB#_&rg)cjKV_XSsKx&6sO=AKSkM$%DQL--t{Ti^)?T=w zyzm$-pNA|^nTGj`;&3dJqD~-$n;m>zRcC6e4j>Ttq=5MiI@74L*@yqhjE7tS3{_V<3Br z>4DufQRlx*|NJs#sA|h?e_o87mWG3pp?0&@J4yH6rm~d4&7xs`cN3+q89cgdElRez zI(T>T9vC5ZfMr6MTQ+735yo!bY5+@_6@@n3W;Sfbgo>B?wathONXLh#D+}3PCn))X z(abIEg#H^W?)eSDj|4nA)O=?mv2Eb8@kZ5xRtkB$ZLoDq0h8p@*V~iQaWqWjEc{)p zc@36a#IGzZ&NfgEGfm{3v*l~INI*Df)U}H@i#h!l?u+g694vqItV1Z$O;8M54LU~e zxp-zVqDn%5SVukr#oh1k_Rc4-Mj@GociAkN;?qJz(OECPneh~wF#Lz8vqoO}B%tWb zw|76j``^(teNF7C0_yvIB%o-^sA<{GLO2L$VzxUSnQD&8Q;b#YAm4-@eYCkmYm?Bx zm{uYl)xlKTXu`1X80+&w>Ik2cu7(|AyjEnwp*S@ZWSI;-(?u;csFO}8FM8z$`j@h! z20L9UHe$DK?Dauiw&{1hB`TP_eWvP_mg%OBZX)LzP)+<&2qz&7s0K`!p+Esowp@crM2zrNlATwl56K$T68H9U zy$uKdyj8V~Et^iSp#A~pLV>d8eec`)y>5vYA8{cgMM*E)$eU)jSJ2QEQg0G=h|vng zm_lpLaVD#nXrYq%mpi7MzCo#moKX4Y?|I5-d(XvALL{!7xMsq~0jmJs6QjJ5)8)te zz0)F^g0xht#A>5N(rzT?WOh!3_IyTHWF{nJ6%S3}2=?Hx+*MeP&MAZj7j)LY8Z1P9 z7gq}r1(PdkSAFsp-=FK4im$GlXWZxO1>c9bW_KrxZWo1q19}Q_J-x7WFgm%-#?NVt}~gWI;KZLYQT++-_Y=p_)@mEF0~|Uv&g{dhh}7Q+w0#e z7*A=B}@55zvfn< z9(=~H&?|xN^5J_vfzDjl`mP}f#ebW>&c6jZQVsPW0ICZFX+n97$>vmj8;h?|s`rEyWUMwgyJ2Kcc1_~!jF-T8Y(|eh6SdSvjuj}M|8KX;t z$a~A8CO|gdrsjDfr?mVvDUWcSe$zUZ3o;+*$r!ola8U8w(TqZx(nd8(Z&em99ZZx;C|I5~!EI8e-u3a_Cdb&$2b64e*`bWm&R`q~XvuE3sB9Un@l@$Y zrKf{VO*gVfSAM$ReSD=&K?U7#j^c0;=J^pHOP}D;D#oxVVc*x^V$j@@lir`m$|r)5 z<-t^cfD3+e^74l=_E6Pfq#;*jirUf@eRu_T8?R2n?%IhIkc8f>5dxD)u;Ggn&Q4p7 zV)~8N&{&CQ@oMoi*4E1a2LHXbtuHb;fxRh)2P|^K7t#gJ94M1*yWPUQkW{D;dI$#eyn$Qg`E)_&q$%Nm-E8K?iXeLlaP$|-z+@v zsRpSpEWQwf+6C-WwtUR0xwhM+IG&n)<4NqE5r^6Y@Bj@U`Y?QS>ytAE=Eq-irH(!x>lUk zad^g_$O()d;7JkOzYk`~rR!~WJ-Wp_s7T@qIKshrjIkW5>R~)5=h|qj>T$-YDph4lj(Ol2&mNw^9sB8Q1f8hp?;K7uLK( z_Y7#WYs}@V@$Tj<5d5LxT>PCZ>aSXsjQ6nZ%b6bq#D;)X^l=cL*LerZLR^)0ib9d; zBml4L(ncKcwi_BePli#q$3^{13n2H%Xymv08$=FS9qL}6xU>}lr)~-_i%ii3D~%o4 z@7qp)@kZoqwSQq8BUv}dWf&agJR+QE#a_GTT9-iS3!z)=c*v4Y8({vC#%YL^vdNoQ zQV%ak)rb_-7-RW=Fo1Xo? z$ahy`YI00Rx`8p6mFlK1WCwk$d5^`EAEDDstd_Z-kDW4)b}I;^kO?wTM?EN z)`o4ugS}K=4$D8XsDgbKM>wq8+f<+Xnoji>D(?+#VH3>Eaf^2p-LolG@bbUOwJO@Y zK^>x+O?VAhx;6v55GnVxLZ``+Wf9g+Fxp5K%eTif`Wc-*Xwy5dfOcF*zwKR-I;L>? z2?8~Kk5sJ~Y-|NtR5Tkbla`ro)!uqjD^(Qhow%D?&D|tKm)-c&NJFyrkIM%+ggD^C zs5FbXl3zGem_pjewn(8Sa}hx34rS1^6M7embhA$;ZN^f2#41u{J)84HBbf_8;6Kj! z0D4QAqx8wP;AQPB26!)>6Mh#S_4UfQCl5C+ywN*_P2w8-`i8^UNzef~k2Rm|$CPA* z2$oUtM4;7Ej!be8`~UhQw|{u5y;~q8U%H|C0ceI8ifaGV^pW^etGLwv@YV6^EB=ge7Xs@c#W}$9R_War6;}edh<+`)26S<2tB-LPpenj!NOYB91yAw zs)eatApGN=>x{*vk_~|9bg6oKhc@87DkoJY0U<+n054i^` z4j6hLNYQ*si&5=3&hMe@T!siTlwy2sAx>*o>z5HkR5<-`7&2B_3}R-bqow8i?p(tB zBoe0_1G!UpBY~TM7S&{gBJCMgl;9W@Y$ZGxqEKjU(qAP57JJ(L{j^N;2J#NVbACE+ zCjWcq(e>!9b`aYmfQXAp0gi-KxM0@XiS*iQh_c3rjp&Xz1br>x6r#Ja?mf1=dpRpT zvdUnHGJqc_D&{vG>{Ht-%2N>vRz{$mTMgXW;+aJnoV%QkE9Q`JczH+Mci+lwS;42P z@cd06UCRN=C5f$GH@y?shy$UrB}-B7vgQcBsiUf*EKd_QMB!96G(Iar0dY>X-0vID z``kJ4yFR)^N0?sgMgcmVhD|5;nR<7`^7bLM0BlS04VHifnP^w#)^_w9SSH@Y%Fg*! zxhdflJ&1>?#t)@95~5{bv?rYcZaK1Fo6Vo7Hd#$QWl}9+Gi-_=Bc|FeI>W!^Y(`>Y zE-z_*9r(LT`7X{$>gd{DVycT$wY*lQ$!tn1wWIyz&S%Q}D&^)Z9bm(~rdw>@tK!b| z7XzK?GSi)T6OItKAmtyZpH@5Sb#q^*w8amkjdiu{dVq0=H|t@&B}x3<;Ql}lQ;{l z2Q@Nf#KD?^t9iB2-@ojl?WRTW$v|0N)xBHj>QdwZ(J}3xXET#_DIvjTmJ7w{ z$)%H5wWGCj=mIOid(Yg|Bd%T=4P5l$xeu>Qyi6JT#B{c zJ~ILxz;O^AI%@pa*|0@r>mMn^3ji%z3Q^?woy#ks95O&6KDqwCZs zN0s=BW=DS`s-ScjvNEoMXY;(2ha>|<#5FdX(27@FAqzXIg`jAGG_ItrNe?7{JzJD5 zr{D&+ViZsuY(1{cXj~$kdomL_%iX?+(WZW4RD+aH*%v0V4))3*+yQ=oSaIRHH0A)R z(E-NLv383U3fPZSOjdSN`-{0jaNa%Gr?A$4pj?cvN zJ#Yft!}G-f>C`g46bvkrZAwekrynw0+fPA=UMna8xro-U!$}xd+VAlJw7?yW@3Y%F z5vF3=pp88SvU@(6=x^m8Pl5KVH9+TvBgcCzfl^lq2Z=*aU`9n=<(}yA3W*MyoRZF6 zTy(&c7Z@-;1h=7d;-if>`^T|1();x)m~ATeyKTfK3Ir*<`w+{h@z6uPL5$02kWWam$R*Iv*F)eT3 zZ%hJpfw;`TV!1B=Q33Fi4}|s3J;@c9X+uXx@z)5in@prcdrhgk=RdFOZ)oU1eIx3G z#v{gHBOmczwO}@Ss(sNm&s1q%gz#W~XESfJ(=CTocWM(LQH|n3yQEndtZNXr^8C_G zR&eIwgaynC)EW0u@3-#fMpq@HByAgx8`}6T$6gW|ygk8Qjy{1(8hEx#*y%*D7YCh_ z;IR{Fr#Cx1EEF^$oHVLOYZRVT#j^T*@A2OgG_qgYjRUgKL2>rj&L<25syZ$Ok>v`u z(ie5q*yaaQUV%l)X*f!nlyDLt(~Ag#UYi>XkW2UmKm}$Nfo7Af z381B06W-a&2;7_UF5K1;FPPx+R}CY{j_I30$|Mj=2HM$PTUtF(-jxbD0G?|{Gfk+g z*&W5wO&Dta0*NyedZ8Qbl90*=g-~IzKgh{;QG;RuRI^WpwWsJ(fQh0c?v;Bttj4*6 z!ay6exOg|x*R~dhgnSViZBJ?+I7G5=83y7#Uy_Hxw0_*+CRP5dcVi2jT(wLVeqKRtXN-@FgH9relfc<606wziAYlmmw|;2Y~6aDD;Jm!{>>U%8vU z4)13=`~4{^EWu;MX>LGXd6NhuN;1sK<>m`=S+-hJr=(0NwM}EL#87GAxOcPC;Vxak z62x9Ct$fyT4bKe>oO^4eXVb!O6(*$UZ4i?@F<5mG4aVwn0>8B?z@2>JO^3y7ndUfN zFJwO3OKEQ>pdOZK#OKM;fSI`t(%Fhsu`!riA88_P3qdS;jzh9qY48gSy=X})G+fKP z3yTs_Tm`jX{zae;fp<1NDP!cu-RSY`W#j<|3VoaW+bAB54>{KUR~VVDoGHxRk$oEIs+{9orSVUR)^j>{Jlz>x#u}2N z+8(_uz_k-Tw&maqYi=Ecwg4JGX^K?i%9K74POx2@(SYuW4$hXtVx$-0DtJ!uTA}1a z@V8Ri)d=t{s0~rG^ICUvsQaL{!)#M95a^~ly@@?nfm6{eh&KZBQ?Rrc?pRTShzP3ZC7)9TL znI$8u_6d~Mit1ie26-R7W%u0?c*@Afb+UljQd74|x|KlK#F@WV50#;VkAC()uvt^4 zFP=dLky+ISs-WPZ$0^l1t*sWwqC%eQ3)Xt0hYwk{tT6#KG_E2)vJR$wHt+Xk#it;Y6et{{3sep8?+ zSFm&CF1eA;v+(LEP$5^zy7lz5SrBPpoQ+4>uit9Y?Q<`>8lU&gB<0ixfgU>sM_Fttr~Mg2a1(8=_MeKgNNZ+&aUHW zQYxy4t!18R&)jfWX$5h*aAL2Ok6u$)M&T011HSY67UUzZtk5^B5JEvrf$m5vug?qa z8ZWW#wTsR$E=VLnuc@6-h~%vcnr5)s^B0IU^1}r^punrJ+kqZBeTn)4rYjoBXpfyF zc4fRt9uMQd@K#~KfycsaaJjiHzBm^__ulN+9ynYL4rcAaXsNQ_kC2n&stzbz#089c zlPl6`dDVYfi_ka16t$#b8@O&2=NL*OUO}P6H<~q1uIyU4f0Lv*wCdor8V{gJ{q2wp zQZGed@XQF&ce`j0CHgL&{ev7&=O5Z&yMF83BY=AHtko~`fGyRN#;%5b>2|Gg^zB;f zk%PAZc1z4peS?ubCJ?b3;*C?*pb1J$M`6te9N1CIG6`|j)VkK%HWXnd^TJBf|8;wu zd4Nys4;R=A(%Qysf0p<0??sv|a*8eFR|_dac(;=vpEOF5t6@lSg7JPKP7+(z>@KV` zXffvSS*u!xjeN|tRY~nUnk=0%0MXAc{I1Ghv3@RlK-85wNaA-tNOk+ z53>7K8K!4FOT1^zh^RYPhhm&DMQ)0Gh?pUDNxZrN`9Z-$d=~GtB-svR^`5+wxwg6{ zcb#rlW|oeELJqY*##MbGAwmlQjdAe6G4ihozVSXUhSsK@TL0X@Am69H=%oiX|F0aO zZpn)B*tCKdd3TjE-!1U2Ip6t_{e$v~?&RA?F@!=s3w_c6Rzo#KSqnv_3Z+kc5ldID z_|-mP9G`ry+v+~p6yD(Oe@S)iBQ_%2UtFW)2l{c5nb|LSidDSh8%5BG|77ZXD8c{9 zlC>y?-cCog3lu`@LZrrgDyzY1|1cjO{YzlB6vIb5(6hQL5HnW~D9TiEH z`V$s%|4UC%#9oPHbH0V4AJlhJne>=cJ+@4PRgrD5OPtR>K{yNf-Ym+C&NM&WuSj5= zJ!5eu*hAEwcVbI9Hbg?+MAir~Lt}6jli0lo>FlOXQC5=c%prXFKEfO~Pz3Xz0anic z4MT~VSSh2+lMvhCNtI(CU(>kukTN_gte#O1+42&w^7{TSG)IF)nrxdR~IGT zM>7Uz5MUTWbKw%v2yUKe{&rJz>>ZH$QACwglJZ5*k?t2qBC$sFZT4@UroNCsqK@%C z22(4(P!pu{7ZbSGd}~ZZ{eTiGrY$Pb^vFkSH%7)gxNN}Xk$Wcs;X6BjQhq4BjL~BZ z+lSl@xV1qH8DC($O$H>$h`%UAUGV))vynyS?xwkLqVRFGQ5L+3#p{s6aVh5e8 zFywcIYYvC^AYD_-EXdL~j+%+WRKMiWP4Q)*I;~ko>$(9*BDOY{ndOOnYwtxgQ5Wq$-pAviH3b}gOGm4k3)AK{R%x=a?ShS z9JsJnZFcptt+5fH7UUHZcVtJ z&qqkbzQWN-V_Ubywgo`y_pob~qhq2%*k<+QG-dmPbVhSPgx31WMs&(p^k^{md&57# z-x;tiDU#H}#)_P1fbVFCZSe9z3KaSBb$l^>4)lhhduKebThxe({-MZB{chM0(Vy5; zHD*%GuqI<=Wyf`nw^#h6wu5A8oBVJqd-<^qF3pm($PlKbhO&vyta)#^N=M*>H#dwP3^cUnvJ@DzRCzy&# z3x3)i8V_Rx9B3aZb#11!sJqX4SbD6$B-6k>!`7uJ!iR=R3r@tpb7q}>?d1_G9$!c{ zr17jrt}L#uoZjgy_NTV272WQOO}R+p9Bm)&%eh5YjKnLpHTKs@%Z0MQy6 zF8m{mj7fKcM8AxVLu)++aMSLl(4>6p3wO^Hg$Q97yRa;9Cdf8?5ZEmEs}Pw(mczp# z9?kfzQk?s}2rnQ#C+cI=Qojsc*QM%-!%CM*Nh5FCRL-Z%HDh*6)xqIIKKgfM)^c>C zms8QlJu?im2WmX3>4`SrY$rV97;}HR%%FBh88 zfgwKdng{2PtCmn=&d4Zo8uUg6KgTq@Bb8rJE0;${#L$-oShx}t9z`z+P)_XJm?SKA1C=lo824)WI7%O` zuVK_=?tGPleD`h#(EL31Rsp*0jE6$ne4+XqHj2qizf827)6n>AAy5pjdU-?M7_E9- zR|L=Ic|Omo>^bm>1rXsMZ^9bXJ=DLFGX~kA1vy$>lKCy3u2}rpqrz@hWWHE8I!kq; z(2;&}+BtJE4Ae0; z?f+yKA`pa)18o^l`n_ARgi}E-znj<|JKGfZ*c!ATL=fz)kHwj1L04MPNw1Ao46`U; zt!$g0vF83rEYTJ;YU9A z-AC+*-uxcv*v(>eVk3?01((c#Mx7jyOs9HmD<)^f!9mKW3d8as9}H4E%_*IonNFYX zqGhJE_XC~(3l~iEw%n^VPu}YW3Y%FNH{o5AnU3>j^mm-lbA-U``AMq+!8_c_Dm}ev z?nIL9o#?Z>N3k~7=qb|**9!w1wr0+b;~O(|%#?08Rv>*{xC@b+wLS-j^a0(ChdQ8V z8PME?P-eGOr%Ly8&Ol#m#fC!2WuuC*HHOWjqGG$`&mIUHFQB;h8D82EN z2~m?|JTTam>S~0<1_;^hGpVaC#j4NycO?GHY(IJt_RVZIjIrZoe9b`c%1-u3bq?tK zUgPL{>0vg~arFFZi;!;U*1|24fwf5QESqcjeN!MreRnk;pW0@VcPvjkJ)0f75A?zg z!kyiP+2wB=EDugB3so4(`$P~Q z9$_TPP4=*PGSrZY87x%3O9W{ziSlY3_Ht?lHb?W(sa)XhXi$3&hO*DV1hPG=l$X}H zIcPS&*I#s{s3J8IxjW_CpwwHQfcJ{@-)uxGTSLopO7>ifKrNG_wYYd%re)#NMDg+WUiv%>Zn+pR4V%T))Qs?g}_36zmPt zc=DU!hgtBZiGW-R=lppD(n8o){pVP$R}l9|FQl3|THb0?+`CNr%W6Qsy;{>@3^T)O zg9RBRfS1$L=dI7I=%D2q*cC3yfSfKe27d2UM|N0_2b|mqU{MpYZ)LVl{hPIVrsksV zndxSCQyyk~uMQ0H6?vf+wN*8>bv z>2(~l@Y#jSCj;y`QB;LG?Kt1ysf&+2UXrxkI6}#6Y*i77B92$HHg^lcF*cBR8AAb- z>E*9xMLegrE0c&4c=joUO88-81n(?Uu{TmG6l59U8=Q?vjl|RTyByFHHX~@9(rLI< zU1!-FLICIVvy=|>PKA5~zo`q7o{lHA<;_FwRAcQ3zqnEO@3UN;rlkD**?DfA!{Jva zBWL*x<$0OG62XN~NL~J2HyR3a%pH9uIv4rMo4ZCHhH=MU9I@{+(WfNFx>Ze1vtW}|vJGjhrJ4wLEpNU2*n14P=_^DTKe`cu92cj8LJ(hc^@l&d)go>V|k3yQu;n9g=d??Bg zcKqWgCkw1t$Z~Q|)BL+rYOCEc)0XU1tP`x(&-Z($+22w422h?fbCUh{#8~u03n6!f z?$b+vMMNbJR~|Ci4M30_u+7My$t_A4UG%(nw()!Piq}g-&y9*QmydmKe{mW(CHPCr z|A9Ur;}0amoj!v4<|B|^y4ei-=-n?`$oUmJPm)ScfxYMGE9n13>9h^PIH*+oD1KTn%PsgO^!wByFdE4*LQ zBUrJQY&>j3Z2nXbD)W1$A|a#sF(xcXi)M(G-VZy8s+>cq9*;h_9h0z0P$OMcMDaL8 zzft<+)#O|`TKMnBtyJ=YfkD47?O_+JY&8Y27_vpWG{M9hnEc(wfqC4Yz0s5DUwN{i z^+8D_F*J10AJBdcas&7Or1=M0r*B6yo;pchMO&yOMOP0XU@VE^K6zL0u%Wu^!+q3= zri;*h8~)mO_q=HH`$##NeyQbrR4-}Ex#$H|gAjMj zYoMBpyuN9z@P*P4A{|HKVjt1O;c_#1V(7PMo?RfI2I&!~);-__?FK0ofUv6M+YJa zR;DLhPR&gqdUErR`Na8bGzEdv$D~hr|ESt_(4fP&)4{IZXt=L7Y~%GGWnrcul1?v2 zAO+5}y(-)GvO-WtE5n>zE1#Y+g0%75Yc_=c2Q5I-zt=~%LFh+y9l|B<^q6O}BwgBP zCU~H7&)z_BxXFV|##B7Xi=>ZGuP-I`#U@!i#fh)#W^_z2$hU(QGH6 zIYwN%KYu)8Tj1!{Z{>V^%Zr^x-wSre1S|xvsJMaK+rK;^l7h$_V&Jqap3N}&I%mnHA6MK)tg)|kh$zz<|q7c-@Cte+^Laf7Sb;}Wv#HFvnppOba z9}tK*&28@(g@|1LShrWzBN#?(GE7qgJUnVI0ValsxGalru}O41Js^aQzDku%_k zmh`B+qPCzhATfTUpO6B6;Hu2_;O8mNpI&(g?qBJx-g@GuWsJ-5tvP4YR0Y`%3oO87 z(V&XbJ=9mYEjHjR-}X4(aoCs0l%-Qrlh0wa_h()1Iwww15xB>|qQCB4?L6m=*_Q0X z4^>x8{<)0v`w}XSCAg4JnLT0&k}1USG#mYHdhNCbJ{J+(Z27J?Hi^|^%Pm4QOYIF> z9)Xep0-rZva#Rb$yaj&foGos#=hShFveM;_MQ%Y`o(yiRt}hbA$7}F&1`}F$F@ln2 z>(M4OECr4A32z^UGsympxDFJxthIl>p(p?R@IHFzKac{~@)mPCen-mU@Fb}#=f8!V zuqfgFLGH71=)MaCfy!Xc<&!Z;uHcu>_C7@S-F;R)bN;gd{|8}2)!SH|!N|inEapKX z>N+yf%R2tn#Mm5^j~C7{E=r%KFIJpuayCq9>so^6Mt5JdMr1Nw>!fdzB7nc6-cVK?Sh8Qren3u&r@s)=p?#dx!VuNx zSr72@cp`74&r=WTSc;bji<1>-^NtO+`SSwX<`d0)91LJW$MIccv-O#+UUimv0O>BnC1iR2u280X8snAAL3sfU@9vi`f=YmF*rb{I z;H?Mi(7|1QZ0de}7P0`Sg|nFuQ*lpBY(}K;9DmL`K(*^Bi&vc_m<#+UAGg_Tt6Z$stcn%^y$?zz74>G2oHCy@CUP1K&KOX!>OZp71jSL~6 zKIlGJ*1^>=*QHyu=^lZQgxM2oATI6R&AkAC1N2z;L6WuCzd#CXR{^0d=aqA`?tZW! zPr(<&U}e>k;kn&KV8D?ES*!Ne8X>Qe-%Wd#)nupb-*N+=zA9F5GSqDANJI3-Nhg~c znF0KMGNL@7>r0BdP;So90LP1q1GYCF9^g#ExPD;PY(nc96gipAvXWrj;k?%x>FRd= zg+W%u6)OjB@_fxqnp)gOPy2DpiHo%qy z`)|t}@&nZ4`6Ph@Ka7a&7b`HWCJG8d2i-5OXO9(E>Z7 z!tNg+C6Gl{Bv3bf`E3E(<%ibsn>d7MXN<#EWnw|8Mkq}mzC*`^aX>0Oi>+CnngH>r z-qmU2x(iLZ@aSi0^=bbYfE7(=s~1)xXqmSZf>lriNEh+m;;)4@!qGd%wLf_JalBAE zh$+?6S<{>W<`;_y8>Kpp)UauA1r*qU)cVet!b}2%X2}lHovz)u9ftRdvKQl$OtTe4 zkxy+qy0SOP&Ao2;|HUzOF{iz44qbobB@_Sn_Kc*Bf<4`| ze+`eyzjR5Y*_F~jzG;@@#%%VGPoMQqMilNkh3VJ<2#bmB)gHlJ>`o}_NxVM@nl|;^ zvIBbKMwE?He9w?OXz)xI-7OAC#Q?SibrDT2(8xyv^8}CmJhI=_O|+HZzmUdTE^Ker zqCF{uasKQvuBY{b&S*yxbHi7?v3C0`pT_5BXnfrspcl5#qsj_%wL>2LJOJco+l_4R z8D~3ko(_Z5*RO8L^e~B_X7VuA`RpnHQloB(l%+3+zLMW<#XW)a{ADj?GAg2B!&wKN z7(>QLj1r+8^EVkT+#8LTgVf~G6P21f{h}OB-P6zI+gF4}g^a{iJ<0zC61g(=>!3$l z|4XRs89C;We^+PIZ05_^X={WaG4J5D)L_w7tIq2w1r^5d0`^__1r@wD-k_l~1;>6? z7IMwEN!TEOHBacHij6jDR5nPuL*7t05v z%pgZ5>B{83QIGfBgjTuB0e;mWQc80IRUEG4r5tt z&YQMFPRXv3ctcuL9%!9w5xV4nT7nO5R%b066woWxMiTeMh(daA?=uAt8Qf(vggi_- zeZ*MqcszR^(W+L5pCQWbyPg3gTOdT-tQgF+rxc~DPNjFpE+Jy2bI7PJKq>C9;QDWH zgKu_sKb-u#gNcNzLRJNu?u=Oo1A3w<>#QCwxMKC}$0*KBq1wl>2H1w%`%&+-C$tK+Uq@&$LMOLV{ONvU2_&_yLGYz5NAEM7^+o(sCBU*lCuQCBIth2OUVkt?At=$ZK=r0LHgm8i}IfM&Da zCBM~w{`^D$isl8XV$Y{pli;N5vezAS%aL2-O5Sd!eWGj0Jbh)(u*Q169#(p@FX0jL zT7ggQwPrQeRJWU$R=!ukLahL7?>H^FZ23}fO$6#*fNtv&Ps=HXt>B-drf14!Zm^M2 z;JZ&0o&hlMk@H-H*;VG$C*bz8z<5|G@T49|7@nGH@OQYc`iD01MMWX;|{R?L?fG3dg zG^4fPt)voQlK8}f7mXVrdApY;ROydx*}E%JCvej`EJ0+`^?g50%D+c*MTGOsfj-c* zRsjZyeN+$~dsXK=^!?I9+tv?k%6~U1N@N98Fp0#`&7yh6=Gvw$F!LrZKN5?T zZ1`cymxEuxVko3BuzcCtS|hT5*jk=X>kSrn5`#5C$o_1n%D`W6f7~PoL)c-w-!TY&K!nUKyCD!R zcRp1ey5>Bkv2l5}B;~|RPBsZUg1C)4G*D5zuRC>wKoc?3(tXn`F?m>2@ZcufFShmI zN@Adk6axedvL@UbaDQda8EbnZLy+QW0w9mQTI(;FG^ckPK6KyKo!!c38-L4%`pC9j zO-LcwaR-(Fk%2y4Hn4#!_Tk#d`3SAeP4-&?J92 zdF<#;q+UO5k@^CA#I78;!2z8JRh)n%m-PhrC3-XaITz?9#WJ1Q^x1Pm5&AT$*msoK9K%2V}cj zr?y9q9$#Yt?n7!pn2&6pLoRR(s~sJNSi>Ka#+bvwJ!`g$U8Q5vtPbdPJt$%bXhM}1 z&;VZc`qg(!_M%ve9Y>ZX6#>ZY&!gNDjcfl<+sgnh@jS@hScXZhUTI?2x=T=IUGv5A4ReKi8=xi0W*lqCD4H@ zvCr@e#ahd`x`3RFio;P!pcG2O%_5sv>riEP^Q!Y;OOUcJL?MD2zN(PE&8mP(@6LyaGZLKb9Ug8ar`1viJr^+EVTQ#G=t zI2;3DyG9W%>tYLM^s9eC8d&$~k6Q#3L`+8)H9D-E{ z8A$?OWJSreKOZ4~d2Qo?X-T2sVa1m|OLnptU@%IhRIP*4@uc!77a{^bNo^#{>?y&U zdn9{FYMw2(h&=(V4&)?Jn$B!-6zCEl!|!A(NyF%}>>mcwPROSBZ%w%45s66<&SN*E zK1qnrPSP1Ua+KWek%?O|mqi3)^WACad5_f9DUW&u0E)(3x=mU4Fz+s9+HLnK1mVkTC@VWB=5oAAEygJbA&{kLIXm1Mpe!g?7-AO1};q zBr-9Oq@GbYNZx=3Q{VJKY+eO~lo$pYR9xYpx1tk78Q%cA*be=Zj zLZ3W#$vRY%Q3Iz1jRjQJYUPjZkH6_k0CJ2^oK^nWk%(1|fUm97WhwPN3=O9w{U&(?zF?Q@LQRn<&>wVeBuJI5Fc}zGq`sXD$D}KF6@?Abn;i= zV05j^nTzp71BptoQK@2VmE-coX<$9HPJ#kCo_R`idR8GT=hk(PnUsN0Mgi7L!fI5T zG=~)-l4OY0gTs8OD*8iB8YY7T9FQSJ?FB8ykRb``M@(1D$m%OCzpz0d>HlLmxxJDC zCYIDb)HrfeBYxTPMQ%r6uymvIcyQ6U9tE1&xy z(9OAJvaeG?)ZJ?qrrN0>rgf9oPn8nH=u;dxv8+9>QS=6@Nd#_T>%LxyAvYufU#;`(DTI>s%Sp09qO@JVj;U$s_AA3kr$Dv z7N>QjDnWW8L$6*C2c`3O%LQDVvql;?O*nT4Na6iF6Ni7c;)(n6LB$!MPQ?kFn)eE* zRo#oUw)fIzL0V%f!0NQAu4@3!IdY7niI0Xj*JR?9~hG3XCf7q^iW)A;S zIXBkMr>0lIG;QD9>^*$`W5&5uMt~vm`I;?}fi|hUs4Nrl*1W4&shY&4#^(XBf6C$9 z$P855BOQEfhEh^Do51KGF<`mae#j8{yotH6c>CWv7^!OQ)cC5@z?n)vebLNM45els z>VK|I{jZB3aFlS}E!nyo4w;J4(`k=5^A*Tos&?ds^6ekL0vsLP>GTr`N{s=Pu5y&$q7{j+o$JZR&CjB* zwcv>Za>W3Z=>w+Dpy1Y+-ZHCAXbuA6{+f$Lopd!HV#l4q?#$7`Sq-zV0ayh<9i-s*#C6@$2Dc8TQq(Svu?tuQEy2CdP1Uf~>G7 z#|{5*A((@PCewTDWb`3@Fw>T0?x(70dVMxa^k7rv*5h_084TER*+Wv*%A+570^i%%S)HF=x2z_U0p%Qjg(wGp zQWnZXmMz<06*pd@q65zCzHLdtN$O3yj?qi>(Bk?yklcVJmmVBlnvFe%)d1d(s4PqC zk{H5mj*$f)Fur&D)}67{Y|ZiLjTdY#9q>RvKay;C>7|{_D`hn-fVHxc3+WTh0^bKG z-5?mF`XL>>QhTW8dFeQ#)ahG<_!;>PiKHtwNvf1aQ|%o0`i>^hcFR3^y2DX(sIKY0 zm0mtwAKl$}`<2w=r6n22b9=KW4X>-U4&Au_oqneABez<`AoddaKV;NN^Q=*Io=)1u z*~uqdixGM=k84I>vP&>m-A}iBFV#U@F{rCn8S>MyVC99G6p#;cH~T~^Rl}aAyq#di zpW_qV$K0bQFAQMD{Vf0oA6rY}5uN61!(>3O#wneeZSs6Rm{DJBz`pJ-7p(iWnxdCh z9oubc3dP-{A!Wq&F75CfGL?ckU2aHv!4b4z9tGrz2m- zujs;O7GWLX$qUQ~G|6CvChYiKWU!hfxMQPqDCtZsJe+^PUU>c9E~6{?@m+}_CKYDk z8krSY!GSmt?8&j#ae}hel-EKdo0T7fTimcRP;j!Ot^cRKj7NpTv^B8qw0|B}1Sa@_ z8<7M*a0wHG2K^*3`OhOxGnSKCkym-+lsJmRR<-tA^g1l;TE7(j5 zy#Nnhe$~B;KI71~V_`1Ef;{$IggshjGa#ADimRp4#t-ppO>%i~(PK!c7CW#jI4+&A zB?c6S8`!`B*oGLEn^-xLD$mY%iNy=Kl~w`r2R6Vf6zbHkRoO@c6U;dD^@o;CWk$8t zm{K~RrM+-e7P^8q@XcR;m50DetF>8j0vNN`S2gH=TWCw&t}_5L%DnM9V_o)k@w(=? zk`#O5=Kbcdg~dID6n3%ZI$J?>Qijwv$*K8TcEpM&zb_Np3}95?-z5{M+b|&ZfSzE z&)(zyN1(USSaVWr(tk?K`zMG%c+%vHa6K(@6gssvlHlf0Y>LJSKx9|W(G`ZN_J7Qn zZB=`z0rFR<*lPx%-CMT<`eN3EGsN7G$AOMsNs%ICW|xP0jq=0jSHa;nB1m#F;zE^EkmgKj4?3dV{dJ%kWz>=9&^q zEyJ|8t1u;O@yry7cA!9cMkSqYX>dzI0KAb)Dnp%^KU`3IkJ>Gg)28VYE1@+cXeRLz zVV1Uw!recJj+rhBh4i+^!2fP-3pDizaC-bJXelPy#TqDW2OKH5C|dauN9>~TGDG*t8AM2{ zEk}mcC$9B_cM8SLhqF5}vSN0*ew9bG0(?*OA@X;poq-2R#=~2{1W()jyk#zjBu;4G zb5C@?`x2G-9{;#kscbT|)KCA}2`XU0UZwqtAT+?bILHAZd0Lb7yevCQmCy#x2 zT(GvTce6nmQDz~2%_%;jQUA_&+ji_e-MYWGc}y20w1#q@=ar_fR*cMAApsugwe>nh zImfY4zTrx?+Ci-jk1;Lran2i>SlD>Zf;!-gBLkK?8!A2HXMY=yi>M z`5t58PuF>9g(gxl)DFH?_vb;putjVTe;l$<^qs!x6UP(h_Kv*9GNiEp$dCcp)B~CK z;8XrboqW(c)N9-CDhGr^5?_&X5sItgD_~OD5%DUnmlawYc!9&Eja0xE-}Pd2vg%joP+ z9m2mZoeuT~e+Rob+kG0MiXrWiu-k+R0n%5kW(O%#m*Dx^ih8Bv>8teepu0JPI}%=5 zk{&9{&G3#q?K10Kj}Yg=quxKx8=j1M@W$QKAP&fZP{VSkccdRmWM`f!1{@tV7Koe; zCA_VGMwo%rB556ZR3CkyRLF-Np)K?Bf^WZpP|0Xh)k_`amQ#x3s(;w~TzQJ0zr;U? zWBoLM_P+@u-DZz7X`3f7jxX?ja`c5rltEgR(ZwRZvQSXXE{X(Hg<#+TujRBcqz7`; z`jg@ylPh~wN0&Ulz4H`v78s(~Zdbp-hr3Hz3gqGdF0DP37CM_gRT@(x$2r&`aw($g zHKq{)fhSkuf*yg{>4^XO{SN8yL>M7Yzq`|1*&A-o>o_>F7-vFlrOFrVS91--{n_Gc zue>&q6|mZ8H#M~>beDB3AsA0bjllp_X{|FJr^mL zNBTki-0;et+iF_e+F=wXq$~UFG1F$|(DVXFqq9%7Imn|FRq4`Sk!Bbd8ok=!ias2C zLXJ`65Z@ou+`O7yyx5}u=djm|!@YbDboup^2KujMX z=}Gnj7{X~Td`x4^V|+ZBga5ixC+;|!VRxHfH zIK&j}(5peTT6=_yWF4CS4p5?Hdz-tGsf-CI4InZzGb zQbkhmIql#;LmON@jT(qx>$}^SB+r{9>k!Gz<^_i>qZCQE@(D`$h6qAyLX9h27sS>3 z^6r#jL_0}lV0EDTBRS{eb4-t6hTgWX zR;E}5IWm#JP`P1qYnn6W!3tgW$@8#uui-SWm#)IHXW3Vyr4F?O*zFZvH)($P$gLAz z5PP~pv=n#XIu;_u@B#8`4*S!*7UX48hbUZ>o9U8TD}u&3wudde#`3%;v-L$oBS=s+ zBRc>Gw!<-F|m|16q zWqrZP8&_hCiYI72O;$8d47a%~b^skIXdQa(7ljHrzI6UE5Qa(G$#s07rBt`NYPs)1$f4>>cP zhD3XNR?L*)(VOklx7kK=PO~$p4oqNPA(ExbjINI=ffvEx7#vI}4HZ$cXT^c)~At1YRCbz{p@hn;Y#-4+%Bb z@d;Vl=Sv0%o%b{e{qD^ccBmhg~V0F1nI*^stkO+cZiG+s1{l_5%1;z>AXh} z4K1Eb3${q5j&FX@d9;$vOvXKT_?Un9LQ4KH%Nbj^{z=_l;m-vC_%?eRLqti z6ArajUDgg4LP+XZdK&k}TE!Uuwt)G+15vwd=)Tq}kNdR7&)QCy5G%{vY3XdtF1sL& z5_C>}E6yvI7%;^Mtd50P6Pdk6HG)x=TnOn^*6~WvO?0&q z>HX;QBkPh3b3+4L8IvLns@&sn?7lN>5hG?Ck~SJ>yQAs7J6b6;ty`J_xRYE;AR?Xm za1?ucXkCM9uZyJ8|J8IXye=zs( zRUbrPe}`xXWJHlMQxLVxxmVRLw46m9Cu}ujeBnnFv-Y8`U}`tD&O9zE?ZC5gyQ)V* z$}2jc*YSY895Z4#6~T>LgoB}SQ<=hVmzcTB(ElG4BQ2T?cNF(;RTgk|>SVhdeTl>N z^N`D)>iMuEqMq-#3HcYLOScaRfpJLoY>$yKKc-9{>1Vb)9}-FR(oU~+z5$+Le}TBQ zxiYNIezzT7gUGSPB9BY%FUr1w&sBaNURjh(iI4i|;t%-Xp1D;Uu>4%qD9O1y+jNX^ zb4xp%k7U)5x?|-ij*IZSmfS7z*IekRZ4d8q_lya(VRFn4r1dwg}dZ>yafKYk|(;mLUbk|yCPV2 zKkd=ZXtbO_KhfkjqY+(U9MFks=xlPyu8E%ER$CK_MhqJ2S0M<;>|N}s5IG4k86qz| zxkI2@Y$!cV!JzHN_NHGf#hphZ8*JN;SuWC z==rycU!{~Z87Lhy{wl8(`YH5s?NFr#D?}HixVrI6?Qq@fBuWk+WqKp5{}7aC%7GNY zXPkDWKE#fXeC_CotBzz~D)}Q8Tj8+k75-=dJGF;x${z_1@;C1ktwgGTse>=3Dgk-b zO8iV&9Z4R^K?&*Fi5}N?a)Quyeo7zaKQ$tvjeG1Tvt&(Oor6pRggjScHs{kzV#0q# zfp`Ndv{o~yjdt#KGJ8GC#T_(o+$;mDM%a>K18Hwxc$azM$}i?FQF3oqN5PGjzYOV{ z_YC{=e{7+NyyW{{Wsbh}H1+^!8R!Wnyf#$W`n+p8Wq*~sRX6k@0qA!Bguh`PWow?~ zjhjAzI?pthp{fd4qC-TRGZ0XoxE19jwUkF*xU3!Xl8;ajaS89~nNTKdq^W}l@~1to z`?dg6=+1iEl9Rke0S<*yE-h`Wd6^d?=wwHVU(WlZZtYYp3HV?H z(GferjlT-V25jNYfrX^*EgL#QW`il}Gs&3O{VF@AZx?~D@KjCxji@07BRO+MNtdho9@e!#%y=)>*8bA)>g)m)UB)k^~QVO&>x?x>?6hzQf9BOj0bos7`Dj z-N__IWdG8KC(15H@P|l1ZO5~#eUjoIxLU9b+6GM__oX@4Q)D&lFsQSkEJrNZw12l) zA8IK~dtO^^&>G3(6;Ftr(OS*3lQss9Yq{++VL=6r*y4u|nZ)rnY}^GMeAkMk7?xL5 z^dpqoB{qi`F$6mpMdiaKV<0kwH|%*R*V-yR-6Lj1Xv33C7xyoLX+}=@QV|LTQ#=a4 zQ{{6>*5Fo!0sHH6=fH1&CT&wvXqF?2M%}CK&8k*1XC?ORxM!aV)4r{oV+poaqvFqj z-G&qKTA)~^R{fd68|Ul`j&iRqDk#)19{hhP4)%bj9ti2_j6V15%OaYW#OaCJd@3|i z|K9Ev72CLV_^pSPE?}n-jKm$Agl#6T3%LYM z?h@N6!?8R7kfUZpb1mghW5AX49uY(x*ZQ2kFwSI?^mujLJ?buq-oPvIhSQPi%C4Uw zzA_}?ElfxA;?g2bPr3OxOBfWhee06Xbp4Sit;0$c)#Ar7mgCNb?}Hfgw@59gJ!{D- zY9HN%gzK6GCJbtA)6xo5Qc<@Va{$PyEE;MzCvZc-a^U(6?s2xQ|(4C*T06~fxycLw}8uP90 z)+aO5E?sGQ=O5n@J}9PUXgKlZKBxB&w`@2g*J{5Gq2UoBFf0kH3KVy((|d;5YZZ_Y z7WPQ5mn19VdG9-!S|&298DP*&g|WvL9Y?m8B)-gid=QxmPqJ?IRoG@cc*EnoJj7#$ z;9rXQq=W>dpw;ZSPlm(yqI#h}B6eG=r;a^a-ff0R4u>oHga@mo;P;NJ8=?wF7Iqih z)1gTG_#V|zgaL<$#_eN-PPc!KBjc7R=S%wTiqc7o*{TgnQ zG;Gawh~Tao*fg&>Rpz>>taX zTK$qk;G`ED!YrQaE6)@qVJk_0W;`G!zQrrA_roH*_yH%1j6+T4fq}Tf@C8yG9V&0j zG0?$pQ6R|{PG5w||7SBnYKQ8)>~OZKX}W{}@qK*t4YbXF8p6u-$YfH6x_u8aa>(2T zx5(MR&ev-Qk}`_CF#W>sfv6?75cXJKmmUG`pPss%L3^WZh~ulI)6TS6ivXmdaSpi7 zOTtGAXO@~4L(#AzYb^hT<}7^`aS}X`R+B8En)6P9$1?z?1-erh5W4`-4T>>msxafk zP``|M=8}ucG`&G|vb3QiUWZf*)V15lkjqVdYF>5vmelD=M81PbW{4W2kZ^{D zAtDc(=U+jd-$L!k>s|GE@1S7ia*A_MIq|E@@DYHrffx5Kj6Y`MA*9AH2_8Z=*e6Q` z+F4bSe<%D85&l_3cMLerszrGwsnk6d7m2`msUpKs@5v$|Z+QajrLxn*tnKE`%ShBw z#qAyXyGoUmUSGLE;o*$`^JMraJ}!EWJ#-e?sTx>oDhVmC{{atbH+cb%y3D<4p45&! z+w`UkM4U5}?9lC;sfkU6J8ec$kG+nIX1fyzH912QaVbi7A)WYNH_ZS^;taRSqT|`b z11~3&n=8Gm;~v^6Q341b0?_@(5s`T#nJJWmAV6^M?6OMf7%&F3xnzVD zGYE#Ko+x*1owm!eJDI$_YgVw<33Syht0#mE9HCE^NMt^GubXGp&z)Fmv@YhtH^6)- z|B*0z8C&i~26hOumZpjN@C{M!!_b?-fkULZ1-G+bF%Sym)vnjSvA2JkXSe3bInwUr zp`@lmcWYh3HL&y$W<%6)j@d^g?S8$)hJko(QBy4YL)b}(@hd~;cY=ZeffqtsTbSk$ z;5&KFJ@4^vh%Mn$NG~8zeH+4T?TZ;1eQ;8mLzGc@7xGxZ`RKDWna3t(gU;Q|itD(G zBUDUIB}A5~QEjU9Ws%CZy^`mZ-oPj*m&A_qNCizW`gHJupV~qOF7TM z98=o*>R;$AGq7}1`5W6HE*$&L2-It-yG^z;WIac(O|3#no_17#|>0)8?fd9W;@(&dAge`WUvw&31Sk@F$t zY~JOE^R! zA%_%5Qj6QMw(%Ac2%#fq_DGF=p;wi|;P$$EeMyrISB;37xN-UFVrrl2%3Jb+tfnTG zV>kyhK`{ddHMdm^&pA&SE1(J4Y5DjDWm}vN5MXJ#t#8K-V#WsX`!YdLFye!aa13a$ zlyFO@Bgwl1B2+)wVg>wvXH?&_tYa$kTRJ~5uv+DUxE!mAuNQ47C8A3zwnapt_k#bu zYMLqzb8aBOlehr?<7diMby@>0i=z2@`WRn&W%W0V)@fo~)r=9tSIzn*cOGV9Ut*Nr zXs6#*Dr-YRkjabBe0lnEMKLH2|9VzOo?{Tq02S%f`Xy*vju|Z_o)+Yr^?~~YRq`pb z8E5WbHC|I!EvK{a%3BsM zx4863rtbtx<>Uw$n9{zOyDEF!$wGSd8B6qtR}_!cL+a9b`-0k!q zE5a9miN5UN2@f@2YUwi9^oRY`{4sl3<0EW2Xn?z1`$-IB{(KO0Wnl|$7oC+vdJcMw zqk9hiFR>da@gSR{V=DpII;)~jK*034*2=9*Pp9R1&&x)`&Qtr~Qn`bq0UdQeV;(o< zD@~>T1xyWjp5rpflltSOXVmEK9CDxcZ-J{caNaY!#oNr*B>SFV_ z%f~5e7s;4#R5cLUgotiZ8);Wkd%l!yUwfAVpCKklB#X%(u0o;(M6l0L)zIJ-q`0mG z%cRI+0UAd=wvkj#C?#EVspn@LefmOBA(ChRqR<0ArOQi{^=*7mP0iv*iJ=+hA6_s( z>;|;oe>UxW2SHOd0uqIxsZnK+GvDV5^re~W(^m+^hdM&vhDabMz|L+FDo~OX!NXJl z;P2AlKT0`f@MC=MY9xEJ_gCEvxScDfE!?RY>B)mJWt0i84FH}WD@!+i-eDXsA(Y!k zg9`n%c2-+`lL?_5`CaCQ4!)~kAn=G5d;;n0tBba|#<@ zR;Li?dc^DP95)8kR2Nn_jzYyo`UJ9nTl@Kzk7ZgWN>x+qR?_nMpExy}gxx*b?Cjrz zb!Bzx-~Jw-q9(^S`ccIQY~Vt?dOD}DC>OC+NfGfEYew4k_c8vIz)iK$p$$^6Csukb znV-GIhmGR$B3vM_#)QKDOTL4s=SI^t*1uT(ZO5S2hMKM?6%bw@8SwcVo_AMa@FP4t ztOH2@&P!Ndn<|uFRBbvWA_}UvuOzY(C^wV1eTmr1J;7*`LoI!H#v-Y+s9e__UFiEul738@H-U0Q&s15&;9%`{PGkmZ} ze3;=zB(W739F_v|UG_S-G?xkOEolVsao6m{>5H$I6vj^qy{vGY9@*@x0$#?au(4aR zN@I>&Vc!ANzWCl}vFbW)mQ=kdLWSF0wz7O}$jQUFUAhe1mYc#{?5PcD!2^V1w16)x zgpm#b{P?@n!ZVcP0)2V3`Ot>zKq97VF8RAI3v2s-Xll5%e;-eMDD^%5EER~E+!VlX zc$Hx)pcr7zcwti?PO%mCZ07 zqutY`YM;FssB8$=Izt?}_-m@(G=818;Y#d1;3R#ozka1gV{zDEU97@@eQt4CnkrL+ zG-B9>{1nD7T9KqcND4ozy|_ovnU6_6ofbOR&*8$>qj#+KSyl#R?U!@i%eIOgFOKV= zIN?v+_6r*pvO7{k^#83rdI*JR>%oxrirZ!|X>aILAfJmi?@rH6CyCDt0Jkepie8Mi ze~!6|fbOsf;Pcd7r5MzCTbKzl_>eK?n1so(8&aT_esmg)1-@}Y)^SQim#3-=?bK>x zNA1@C9gu{O&mwhI%}#Ux5Fox-(T2vaG#m~avJwR7NNBHFR~w`9sQBdQ)a*QARajvb zd1n;B^n-6f{wHhkLlVCLoB}siUjqF_(c7Q;rtn*f1j2&c_#f3irTZdp zrTqNbWZzrxbgmlQJ@e6(kvyg0Dc;lxFgk0`tP=cs94uEjQGAbJaVP5W;wKO3G6A5h-ApKc08;I)@i(J_r9&I1=t*c|XCsQQ@m&#E>C!|S1> zoji;Ne{7Om{a1Va|KXdt`&f5Mm%Vr%EW~m@8Q~k3#nq&gOzx%G_!SK>8wEa_I?5W09C|<}0_mC(R3+o- zrJ_Yg3+kTycvP*pqpDOoHWa6csbhHsvfZcim3xqxVbIZ~e62>QIT`?oOSXky zABm#R*iB@&(pYou=P%ae_FIIa>JO$943zwr#t|S69FidzD^z6I{5MveXHsDm&?FFk!eHM8Fz@K4jkDncmUZB2C4z0 z$Qcpe>X>&K#Z4K)-tGYx>DwAN^xdwA9dMJymXksbTC*`dqJnU(%X$h+NPkq~GY_FC ze{c0f6xlms{vs@&v>eJHt<)?bRv^&pm?cW^#)anOsQLJ-1ClBkN!PnC;p0NID$$eS z&4YAg_7yruz@h=(+v>&?TjZmOls$`Hg0-Tt(XWG=`Y3KIH0t^3$P3^<6;(v~eF3iq zZtpqI^Yk+GfyBD2S$LNX!pNtkMN z0Rdaf&uU*YZXMR*3OR4c#U_1q>r!^bB~s0{>FGMv*2rzd8a#;Q1GtPu4B;xT@;fx_ z($jlDOHme^h}4gf55K!Jki*M#yYsD0ctpIme=^q`>F76bUf|Uw<*3S4wAf&VB|F{I zbTD_BUbInR64k?~qWhO*7Xlb4uYf0F(3No}f#yim{;ngbM=Dsg%fpk)MNlTifVpC~ z(cNdclct#?9?xkz7d`+<{%^5H>ANJ)*ccDyReT(V^bf5Fo953Tx$aMu>eQpWFm2A2 zw_zyZ$hb)YO#E4>0qY4qRR)nMk_t_(Uy_VxJz)d~Q_O^yg0t|CAa1!A-0-fO*$Ser z#&?*#`u~N_)H1uMSylzWr<|yetFx0lIKXfFixC9M+r#Y}ctXeh3BV6SEhy7kRBYQW zj$**$b=LxFMG>k%dvZXcs!gwGHFa(^FZQ;{zl-v`MSGzs=uY{tPr^I30&~~whSo^k z;FT5spzl$cC*~Kx4heUy{#@$gPVS>Oh{xDv|BoxLnBS8Dv?9=pugr3?l+^r^@6}Fp z0Mt8~?*w3lqMA@6)dTZdHAMi)`(XB2t)~Ll6)?JkGJg;yhxprrr8G*8RwA5R!RoRQ zp&iu>xx|9~nL?-pSUSR;Hj6hc z2M_j&L8~CFnoKYQN)j)L&tYn@I#Up(iF|&ox=1&7-ER3G9E57q%aY2*X-S=SPESkj zee-aVOT^8_7rzSmK)2(*($!3a_VW~fi)S1izebno`-kCI;CvQunVMJlk187 zeC4JTzDK}sJBh9DKwNI|jCiI>WG#_N`<;oN)vW+IK*qoDq{@RRfEN(|kefHI@bJ9n zeapSvJFeud_Fs1}$zYWSb59@^0!v|;O@G?Wd~OPSrKVUiRt+qOajJYCv>v%1Z zysT-)Z71dF5ZIr9IRM=gIu=CSBh%m8tmNps<05b3#sCQrl)o@f_|zqUE6YBIhJF#i z>H!|2uZ0)k0775yKB?X^2*EFifWKlrso;E=r|cq~lS~{4wOXD*pBS5Lm?XHAZ(c`!8fOs5dkwlfF_iSXG9DPj@z<7lb_UoK>9S0RSN z+hP50DBpHE5GYg&EKVgeHw0$E??a%GA@(af*!JltuKSR#¸`=(0<_X2Ev3FYD|QeNdG52 zA;aA!USU6a@su@(-8s#|sWJUE@gWq7{e9)(4chAkRHKlk*gto>gKA)~YUPrvNq`2? zerB)N2M3+eSAVO@hPL`gt-kPlSL7^tpJc2i4j~o5YUVBj5&FZeS^|sdi-xS&O3&e& zl~7^D;^s~37K~bZeqbDc-wvxO-+#MO!N#!K61FxW)xmi|t=6r83R*Ct{E+ImVrC`C zTEv*6a+Ct9??5}v2Mlld(*s~<-%h1sC6`lx&oE!%6`2>xTayyl(zEb^9SJzE6;>AGuM`iN5Uj;zNG57b-_D%UJgmDZ|y!4BSk# z!=}KSso_g7f_D^<0?AhQ7R8E)XQRVSSl^t$y5MXrw9}<$*07g`g}H6GSg6|{FZz8? z;4GGK_rI5%GqZV0v`_IC^}T=HRnJP=Nd7%ka9dy@GB~&vj`Y<3t}#ZezGbpIO)^NB zF}^cf&Bg-bBh(k>wJH25JE%2>#RTbPLX>B+uue-2{;f2L(yb=|5VBV0)Ow=vi@gVB zm+v&$^&aD7O0sxlDHV}sKd7?v#eQzaZS%C8Tp7znIz?eKISwEJeEJB;U$PKGaTcSL zDM8|F#~eA<3(E?IIw_49j;XK!u;m^m9hFjqL|)9-d{Cj~jtREmydK_2B>UHE)<$+8og9}S11R_oN>ir?^DBNK%k|HOaf=hqMs z@vR72E&C)7(C7;V8tH|a!u_Wp3E@ilsUmN5P%yIt!?Foy3eI7m0g;~nvu5D&%%hJ(r@EN5@XR$S^5 z)%~yPQ#;d8k+tIe(ct>k1{?r_$%`xTRlI(0M0B#RJ||Ykeq!@gKLa9{j}&aPIGa_i zHmEjWTwCOw|voo(md2{E-wWQVy?r%Vjx%B2RL|fc(Wj-vzib6+X{&;u13TLfmggT- z0Ta{cyXTQg{u*`yFKo+M-N8k5Lqd!Ys9B~VQZh>^kqluo^$Zs!5X7EdmSbPX-$cnC ztPcpDR3vSa6Awul-X{6()pO5E{wsLv`uIISxi!Ij>3vF>yTNrcN56Ij%Y?EL z-CH1+-{mnlRoW8pn?o$shwre%6qIG3L;oPAYi#yO8w}} z_89y)w<#Z(mY7uO=KeATTUaFGU(aAL>TS=~VKz&P1TD@R@)D~dd{qh{%v|%`u1-4l zPi?qLtc)*g-rIecY37*ThBIse`9RZ=HNpmJWpnHXy+)9%0Mx##-2*9Dc6q1*vzO5Ho=?vnglVx*3QF!-_ z&$G2b_|=`nSlHHjOa8hB8SD)18-!`H%=2cj!uZu^Ay3-k)wN;~ZrQAkq5+rcVyhkQ zZ!Z$O)6x9Y9FHM0#e<%K`EYs*NuKYbD%Mu;H`q~Q7%)g5GE?ha=&(bQ&fxZMb>}ZN z{@Yn%cIOcy%kA896q3>l4>p`+g|>y!rIIy`lBNMw>My#ItOm;)VzrSd5y@;1FnjTM z0775lzmWK9ZH6!^CV0?51u%~%Iu4djw)YeZiD3R$xoAJ2dHqn%X>q)*7CxvF8xZ>B zoQ(jgm--=2-DkO;7TE5@U@IggNB%2}5pG_q9q?e1mf0np^@ylIEYXVTCDpw+bF0#u zw~U4e`OV=5ZYJ8I94m|Gj3yCa_d4uxoC7!SyGRx>iGFoxFjIOmS#)JHHfSv4JOSZz zO^jw_>gU6kIlO8(-=>AGuqH?I&z8cj81+*LMx6W|e3) zw^pDiK)2t}^6uc%)fRz4UrIc zzP^a53J&_0<||~T!>$53n=#h4hq8lySiFs+v=o($mr*V)@G?T@7`MISlF~11k)-QI z+k1|s0jOB;l{plV(D4v!9dnt|W2 zhurmFC5oSINrKr7I=A9>UzwS}C?*mD7-wqZbf;`To)T3j#SvKL3SpFBXB`)erMKj; z6~WDVvOx{%y&GCc!hVf=sS6*gTpb?`q$CSJSBq*s6?HW;r~4abH~_uU-PSkm=sBWFrQ+gQ!GMJsP&YGz+shZp@&}4{ zO?ep=rBzSt;n2*Cl?2`5nxoR0eG>%Rb3%Z~OE#@4R$^Oq8eZd?pHe8y7lMk?ASH3I zp)upvV{+V+r>L*rvGhWv<@v04KpQ(pBAwp3kxwHjKMeno!Blx;pH{0rq}#b88SF>) z{8(v0iO4&YAD~T%f-L^xU)Xw84)yn0rVDGpLyZEgdvYPX&kauX0E!ba-xsomdBN_2 zg_23z$(E$#>yM*;m4(171|w7!6kixVGAhO+FCnSwOSiF5TTTpLl(sNjgaehTXuvlo z0pl%TuVj1i5KD=Q-Qel@#YAZKPc3xBcEKD%h+R(Uv0xlJJ6>+u#}N7=0T&u>f@RCa zrdX=uS~PV(A#3Ls5ujP0XeGe_hvSe z{IsNpY4Z)~T|FHqEgS<1n>j=L=$w`1U6J}7(n1X+VoTz|@QWz}MC@f}93PA#I_21m zm>HhsH3lZMB0{pB1%bO8By@t9b0_ckf2Vw!MIsSh{(fm0nxM%72kx~Ou`u8><(Vo1 zY2v2F+%ZLUuo41 z4)ph^xohNL-#8ZtXCp3sZ3sTHya;vCz=kfOSekI`&`$Tft{mr^{PUKKZCdK`lW8!`UwrL1#^| zPSY6W4Np_6J!a@?y}3#EU1$JlCfI963XvScezO&4YdE|A_}4&s^4wYTO?!5N(a2-F(W{wV?vW? zl>?)ji|ua*1M*QODzKk-GdK@ZetM&smX3}$CH+8O=-J*EOzMGqq5rs2qXj8mKSc?r zBfnQ)w5Y?I+ZLe%UWt>$wMIs=4D6yar1gi`>*aAHjSh4BjUC}%bgaQSkK3*2NO>lb zpW(~`#9)49jo8bDMgm9<;BXWln|B@Nyip)o^4%1F?7RDJjIW%Yoy4j+iXwMi*#uG) zP3vfUNW*eU2A9*Hu6h{zm3Y5`yO;rkdfA1?pCM`|V- z$a3%A+Z}(jf;_|gl({km9t{6YEOC7wit8f>aB8gZV$|RNhJXLyIJ${mh&Zghwr>Jb zU`z+nbwJ(5@oLhOM1RaF!tXQ7evvl2Df`d39lzWV<~a;bcBEM8ZtbDun>sLV)eAAx zGQLu^3N+M}=l*H1-=YC>yI{v&^4`R7>5#%K>2SMofkBX=yg4r^myMKs5W1`n2^m{g z=cArBFIG57O$$ z_N9FXEV<^G|D@u{BcE$*%%|$mXj}Aj{?1t1#yUlA$UshV_2DF zdUdtCP*X^Fhbz?3lD|Rhhk6jC_Pv0?NNV4_SgX0?eYa9`TkbGC`ls^$gbV{1IumDi||ZM5v3iKmeP8x4OripfK!OP@If-mDa#gOJGLL()lv#6 zv&PRyrvIjDA#`xgwh5rj{%7e~cG86}p(IY|lUZS}$a6YMRMxKfh$Q1# z`>kE7J7MmCJ`@j%l(&i~!tiX32&PE&3&}R~Dh4=6AHaV>bzUn%iYp#Y*<7qPFRV#D ze&o+uOt{VA`+7;Xw1eZ6O!`P2{__Fx5E~e7%u<9(%R?7VyuQV+NZ940bsG@s(u-0G zKy2R!(7g$7vS^r=X^bfoqE=Oq43w+!rG8p}$lAB#Ym_1zSD3*-#ou>FPbt|Tm z`k&fQ_62dIH);f&Y5r*9a8;31bmi)Jp^k~YrC)66^U!$=_+__m%`&Y8>&SY!=BYrL z&8i3dl-K3<*a4T7-sl=9eZ~=VSVg{?J5|+eyQ^n3xX9W zKa4Fbf~PagC-N|^6z6A7F4?$YIEy7{DJy7{<36_k`Ru ze2AjmcR-&(&`l-X1>C)FKpoSm5v(G-J(wEzJ`JN_{5K-SH3f$D{t)lP>{A0T=7jq8 z5iIo=yu@EDl(Wjzx%^92+N4loC0mC0uf`aPgo0M){{FIbv-6&vczZZn~ zS%HUf^HPHA zQP};c1^cyDb69(ZtC{$`7v+ZB8qhUQZ0lVMlzHJtjv8>ZG-L+xV}uil5H(iFYN2ne zjuW6MJq*bba92@`VNzl9{!$jWN=8P2JP0k9k=T!u7@)OXT?V2@yXRUCbm6br3N6`D+)J-Fsh|w^yG?H>=dN zHRi8<4AvD;O8ZQk3<}y|XUm|DP6JV~6#hs|6j zbBf4A*EyJjvZwg%_!4aopZM}O!DM;GwKmS`>WK61#i4%!o`FlC@IvJd;IRG)|gK{HA(S_<8(_g2Z=Cu}LDmkxA!L#_Z!=k90>`sX3(hC@G*jBDv zV^SicoQ|SYX=`q_n<2#X+G$KOmzf2CP-?r3bHwsHCU%!hKNrnacEVMs!SMoBtuWb7Aua}~#HaKL6(7{MJ@)&Km`6%rI zILPuG0gV+)eAoSlO=s4)BW(8XdNEd+)%L|vL0wqDN1v!gx@#E@sqXOx z@&n&MI=k}0hv?#0XJ9<>Dv!g7_g(=|oRg=3=!3tSu~jVF^99`Un~9;WF+$P!(FG=jIdGSRiA}$87LxcH^o=R!b6Us1llJn!1(0Z>nsHsleUPDRp_a=g4^}960|% zaGe5YE`W5Zg7Qve!j~Rba+iK5G>JL*Z9xaCW0><1!pXIs^`u9UPF)viQpNPdiEh9N zw4V=0eZoW_X+&q?_K(TIj>eY7f zGI6;EI6io6<}nnW)CMM}dVOwc2}}@f&aTSkWNYTm|K#O$co>Q@dhZ4NCiPKlBrf0`V@++>^75Br_z^aJ-Z^TU6YGt@_I$qL$c*(q2x-?zjFuQjZ9${ z0qAWb8hyG$6{=vI@U49eIOvW*YRt+vo({aHE^(KeR@}dV>AnnYNlnu(s~?<5cT|T~ z_t^;!j^8RxsbH)(i1`TK#BQQ)^@e&v_!s)hwMJwr@|IBUh#%8z%P(LTL@Rvc@t`}d zrE>c|QMDa65zfzuOjbQRqRj&L?N{7wI1IY^t@R{fg?-gvbvP~F6~65scL*;!quf7z z0#LrEaOm@QUBrB51*fnB`qLy6MS=6KJn`F9W?qF2mkGMyK zsxt7b8zn=u{Cf`c-WHmA+g0=sst2Ezn3dh4As9b6E&xDuBnyx67I|N8IQD3ElF%)e zr|Fwtw4P3b-A$4BQbj+~prc%pOaH0B;dLPytZ{fyiwL-`zEsE4z9LNC&7!y1!&_1X zHiY?U^9ss@a%Fe1gN3t~Jue9Hnc(M7r8uC*3qO%tGi_&_hQOhZJCo=Hy+5&eJS0B> zFB-Oo39f#)TZci@4%Ga1fLZSA8=K9l_^H)nbj)b#cm9)Gk&sr-2tPYSN8-{MxiK`b zEwR!HaTnO+I?Sgt$5>s`3lhHfzJ>>zc8mOWUgLU7qiXqZ+4PrpLae@B8BxA$6h_&< z<#}R|kJZ-VcExoiei4T__DyEjmdQ{wowALb%=-m;m_Mawbgd0gUQ`IBR@$#17gS5` zt^(6hMp)Q-vH7tpPjJNa$wP@Bq`wcQLa_Ml_+B2(CE95Jsxc)q>VSq&1ZZ_bu<7m1 zZmwqC(W9osS(l2?r4eC4p()6Sy?5X*v{;g?UWJnb^$Pild?{98m*zT`Ycq=>R=iS%lPZWEQL%vU(Gx-!VZu5w8; z;3ueI8j2H0%<;vXrB0|qJy*y=<#o?AY7?HZ$se?di1c{ut!fNU^ADCQWqBZ-&f923 zT4)1$@TxuC(#$GSj2{h_+L5X76QkCS?{ef8;bH8GHr_rZCpyn}9L=hlQncede;OjY9T7rZqn@nVbx0 z##t}5@ZtX-zeGcj&0!h!qXPZqV6muu8fHlvr(VjV0z}1ro;xPfz)OnBPX9PPDL1Pc zWq&?Gm_^_pA&_0EMxW*>$xJdD<;#G?m+x*m$I=m)Dar~KfA7^k(})uud6Qm$lK;WH zgSd!zSJI*>r!0x!81|}9qdQi3*i3s0OT-Jw4_Xu%7?hqDb_Qc}x$+=$9qV&K&efKO z0optivap3{d2tOfn2a2dUPjYcUWcvyt1?|B9A*S=~`n}&m zl)St~P}-K-+-skb8;p=ac3)FEL(PC=%IjC7j+&~eix#WkdzH`#>(MoihcQisEzYat zO_Y^N3>b_8khwmL2-Me#eeKa!ufwohNW?59vzcPHqU0eY+1?EJw=l0Yr_7a9Kkkz6mWT+M^rE$@|J9sGz3)^!zT#4y`ub+t3(ag3nsu#|yyc^GeO9FF1X^eNyb8LrI zw^9BxQZJ$6-{#-(x*|!~RijCTbZa+m%@D=u^f}`lDOIP?UKVGXPG&D<(o4_ zME6n6IK^Y%@ce@q+P4&Qrz!_S{V04#4#o?1o_!hbbOEsHW?Zou7YUDIj;EJ7ltH=< zgV-hXE>MT8y)lpMtTfSA;B!+c8&PI~Cm|^`{%TRS^Vm8Uvo(pXW7WAvEKyjow^|wh zm5aVT6m@h+!32h5K0L>3{iSNkJZ%|MCm3A+^Op|grVjQqUp1?CHStaTDinU2NO4hH zN!5|ESvwN$$VDvW-3)9NI}0FvxzmiA@oRV$J$}pO9Z?IZaQ~$9mta_0O!REGgz?B9 zt`8{8{U6=lw7vVEJ}DQZQCU&cIHCiVG^uMu1TaL(qDOvsCP7Jmun3MQM>of8!QE2A zW3sk~74FT0qX$`yh47lDWor7GMzCSZ#DwG{DXNcP(eTxjgmsTw;Ln;?kKo5u~4k5I^XOTt{hYpN9GW!Bu`W95>675(iMzLWA3ONtr9Kn3q*GAyf# z0C(Ix)67UL>}!LXmq!oae{_-rWnd7RwG3nyiIgWwRM9Cx4gCL1jR-4#W`5m<>mX&1 z7sld`!YA@@r&h05SN26=$nhiQ00FDu>&$dwdx-m3Xpa6EZ*mqq=6J1=87L!+ZP%o} zdCL%zuicTb->A2u@9a_we@19H&%uCfS&ImsWu=WnWWIf2)n8H>1`YYgib*vO9@z-s z482uj^!hAv)$rA?zt!c%`Njr#Vgybi5RTBBvnYuahL5E<(P{Bel;>SN#!W$O=MO+; ziQRphYI=Mdwz__s27vh5@;bP(ENy9?Ay{AJ;Eb`+$EdmgE>OV zFh_Z5v?Ld*Arh%Bu*4Df`pzY-GMS0^3Ks>Zqm>3?g*qYvmsMK2Om| zTWLWTBsmcus*q;llHih0F93G$hL%_`I2P4ycqm4}a1^fDPcB-WFiS+C{g(Rlmj#e& z)zE(tp$RM*NVceih}uNF#JW{&!HK|3#B%X51#Hf5SMQzEn z=F7%|jFox1eslA@;!QlljrQi!fw}RgC_>4KCN5VwTl3CWxm11O2iZBtPf-v4U;E z1c37N`rSI6@o^pPZXB%APgzgvvZ^yElLkEx@kTE`;t{08-2ZQMpcavqpd!nhd6hac z;?d%Yh~=b207g|1zFJaR6Nnw%*dtZwR-oXq1p-*{C}4(0x<`N9J8My zr##H!WL(<~^CifTs=sLHw+vY;OfV*U0BooqoPe3mDe@3(^~&RVPjHNm>ES$(_#o?bAr+pG1G)TA2xwc1QtT^)Z5JV}*Ora!yEX(Yb;^T=|)f;aub zh=7V(vg$~03Kz%xE}`7Arbksmr2jRJyb-~Ac}{u?31N*saHekxIlEfJ&|fBeJLG!y zqy}giZJ_Y)-)z`qOT}6NYNDf~$#54hnjMPJM4{Ikrm9CXU|ET!1yo65Lsk>#hFyOR zJ6;^558(k|$Yj6!H~uE?kh#%XU*$3}g;!e)>-jf7Y6 zdI%|_jNR)aZ^GI*vGht$BW~O!WfA(2oSQO|=C48em@Gj;#HZ#m;#%q9UOu3v5A z!h}|VxmH_El`}%3m)|SY=D~|0s2})ekP($>pInYdX$Z)lY+2(3^a+)3`b#SA|I*n& zO>*$O@HM2ErsM)^fW1Ck3$ZA>5(?IA3>b?4hckv2lYOhI z$&SJ(zHFve(FpAVErrN_5+eQ62yWZbl)^}u(<}r)$>`^_+%ecm4;;v{;YQMo*!<=1 zneC3$2HR&j{latLnXo05h==73$kA8zXl4sybWFkJ#4A5|eQuJNUEVjLh(68G$%&N~ zL;;%?-f2t@64L2@js!5f93?mr7=+eSaCO9>{yi)<)vcn|+dTC*ZuE&B z$#+>~!JO#XgR$3Uov7R;5GpxuZTlMEMd-EJ2jF6zZui<^O72rn0`hkLX@~;3yEc@a zo=M8sh%5Gi1FWCsNd6oE78MhD&v15_isPv3nFw**+7SJr+pdzrpGyl8#Cs+BzwoX^ zt`C_q8l8&aYc+#JN3+4!tG@yRe5^>f(173TrDT6xO8lR-&|oJcOJCT!RHNj*iNxkw zJ;Lw>={iJ;!CtXA3ECb=<~Y;0PDheTBMe(898y{!jFhoBYZ?7iWXABNYLi`c(wLI= zspb2y8J4>C9#T(VBOw%8#A``4K^zO@XaaUGWwaw+ zHhf2xus=1po|ji~8!SuE;Q1y)=}n%Y9TIl}6JC?3-1MW72hC&m>-{1(QBG;c?CZ)L z8rEOqj}rR{QRpM=V}WEmHOn<vBB2)^*yYxo1#Zpd2q06)LL<& z_q95$4z?9BX?~6_a#D_u)IMvGR?gcgx^0pxHhbJFb2!a3O$GN7%wjWe7`$u9N7|)( zj`(+iiF&nC{d&E+MCnPePTYd$f@`M`uIU7Z#P#(%Gnn7eI!nx% zYI*E0QCEVY zo+fl1DSi__v3_94W)CmjV}=m{ysug3M2c-QypcI<#gMCC-M)KaRgv@e$qDeY2c}ha z+f;_aEHVP;?wi2z>`cL;|CAp<9({iRppH-{55%1ok0^WJ-;ZaD6v7D?!j1#mS)NKs zy?egQ;YElgE8EA!i0jEq*_f7@?-_-qm)1zV#newB_a!zUlmF+0$86|E*C0(lc9*XY ziabrUqt>MiXWYS<$>@E-ZEVJZz3>q*^VJ9%Aaf6IG@*_s;_BZZ%OvCDPmRM@u^AQO zp<6(;!<`S!@iX9odv#oi^6sFWn9kSY0~s7${$Q8hl#HT@{Fl{%ZDp~clW>sBj`FKd zAGeY1@P_o~1&_qJBkn-13v>y;h9snfc@Y1}%`%f}%VW}H`$WKMT#l0FoT?d$oa&@l z?#W)$H8hwg{+NL$RqSv!1D>Z@frG za4t6^FMU8|ly;+dDs8!C+W*eknuz7stJ*6 z+RkP5Sjov)Y%tFJuc#-n@p5(?og#%InD@KxJbwx1* z=@jG*OS;vsJtV<)4{F2r-o57ZAQv8#sLJ9gD!e{k#I!|NB0Tc+^RZO(Oc?zIk4;d8H(!zp%F$aS83h)BMkGdhYK z1m?wKOrBn@4 zU79aw6^-VvcdXgX;Q1cotElCae>u4T85lvUb5C#rJseWSfX5_fk2UMNicyVAdj^Dl zry+I3FduP+si)FaH*fRG-*`a*X2i?|PQv04vpO(BE#WMlcUuMLc?^z-E8OGrb-0tX ztb*NQHlPH9VK|;tu`boQF2?p+G~T7^)Ud8sU&+CHQg+a_HnDzll$ zuJu`nb|!J#l#=uVBXj6h3{|i{wp*@A7qT#^Z?Mg2k)a|j|7JVdD4W{@HWcGyK}2B{ zI_@wlME5wNn@@mc(=Kf7(L(F6$o|hrEJDl}T9J^?3&C(ld)*p77HT;EL06x2OKW>O zn0Cw)VLC>AkE1Bqywk>98#&v>|97_FClCYfY`uHzmHqKu2}f#cjsEcg+RC=)gqoo0 zJTM3){ne1qM*4l+X!O8L8sU%r3 zJhR5R`55FTc1nkO^SJ1j4$n$_B*${As3(1-HpdBnb9 z+8}-5GhNeOkk7>rE;|J~5jWJD-e}4}rx+!JeC&MkDFixM`+UY-p01!&1Zy(cfrpfc z9n%lbOQ!+W=ZJENH%U~x<0QbJ=!hBr0k)-Bi{$OG&tzIIZ!T#0+?}=%MjrifW(HvX zoJ{I~nsF04_riK1f2I>lo30s;u(D=a;KO&B?>a2)ZXn zdI(t$+yyvM28F%i?8_ucHV1xloS08-eB=dk?!!~sWv_=ym?r;2g+gBWk?<#I*+8Dk zV+6Ql{77~@PHS&TX-W?rTGP%HNdvf$P{PeUZzK!y;s_rYm+8~xi|J(?dcl$%#Orua zCd%gyTz-^I&oO-=OQF?3aE-O~vpHbV^8YRq$Eq6u5~OL)Q2(=?97gQ50~kDJKd3hV zoc{XsqBP6y?48bvb_1;LG z?Crpx$?RkK43x{mSuo`u<92q}b_W@Mr+*Gr&<8r|p&XxmM-^Ujkbqqu;pDztaJ>wS zMXo*_%BozN#mR#KPse$d=Ks8x{s+Gxi<=WTDM!x2mF}Ye2DUW1yokae4m2vbOA6w* zUFUmTe}XH3MK~xH4|cHkN4~RruONxOW;8nj3d)-8IGr@Del>mNp-+T;jJ9lEQh~_V zX{#UUEQtInRaS;P&RJJ)HVAR-boQXsCbhp=qmBbPh<3kyUN}105ry{trJtAwfuuS}K839_EHa8BB0&JWZOZmfKVYj3y=raw{%WWdSTx6F$Ekr1u8v9cXaDum$+EmC3`Y$)- zPSq~5Yd|gLV2V+JRy3bE3;3Z9CS4+&&wvo^LgqY)n70XSqowt*g1_u1G0_tlh5gkf zswSug_-^zw=h={vHg}eQd9M4GI&zr2R)x38mUM7&>(x*B1C9o8SQuIjws(XSN?VoQ0d7%BL%xCQzI zTeY6|Svj)`?!K45xj~1KL8GBbRDM=!oeAF#j`?jxWL^Fpk(Jx#L?E3PbC_F8?Y#DL z`*&-I$ITQF7h}zwX05@*gdIbQ;HHZlO6kgaC7F=E?CLyv*iF^{GUfr~)|VpVi2fNf z5f_|-RC4@n3ThHdKkz%{okJQTQN--`J@U{d(nwHKwXj1CSF*-f{xeg)7x2RRE5QaX zvk%V?4N{Y#I!T~3g))L7EcbOttQQ?ArMOYhxjE>CW0tQBYP&UDV}mf8+&-+W%YGaH z0pp8OD`eM_GBlxgH*`UVZkApPPjd4NyhRL8YatzAJkDF0H)nF?<+1tO%Pf3+LD~OLRd=ish1f;u zp12jk^|$XV3+k&B(#%-LTda>^CY#3a)S#ok9I4Q4BL+cG(R}~`xjATzvvXgCGmn8E zXRs}dV(8u3;10d29o%!q%d-#eh_M*r6Dj7e3S zp}D8IB-+@eirtH*s{8oU?wFfHUcYR5#MBe`76jm`DPd0SGeZ$+d zG}a>MohVynQ=cxU{0+2KT^h0ZCkie(3CsG<)fw%x{q_|=c!+;T9#I}V!}PM+7kf0v z*Zl$j;1L$c!JDC3rxYl`%OOzb*qT6X$y9c%c`wHsP|&B#D(H+JP1`V<2h)j{bbO30tr z^ej0d>X3!LSp}5Ckq9@=9N~3I=n{b}28bDaI1=m(0b&n9K`m=r$4 zW|Mw~&oYw4)fHx^y0?N-7N|iRB?`9_b9q&m&75Y%vaL}KQn&bF87r$mwL}mrNk%LX zt9nqj6bpgWriYAK+0f}{_GeHduTv^mO2)8d0v)2Ttez+@lKwJkwjt5{ey)L0xTf@< z|52l5+YQ4c#4bHsz1;^>MPaa^aRN%cIGterF0b>Ah}Z3dA0 zkuSh0W|PSMT158vroV%$W0^C?y`+K;z4P@DZY@r~e1#)(s*vb>8oOX7Q^>c5YPg*m z()a4Y3R^24>!eYE9|KI6+Jv6f#?*ACDa)qyI@Q7}^!0I7qa?A>ib*;w*U`Gu#dxdi z=n})ta8MsbTu#9Yo1^wr^&hPfwFaEla;U#5W~+8#QVE)r95wZu*23$oA$K2p;VKH4v7HmW-(|$ac05c zX7CgZiJaz}jKvx`TtTbwucW*A+*)P9VH#j7C?pxd-$VWzM-h2^mDtQ&GPF zM6@;TndH;`M?$UpW2ZBqfOghz3zcw z`oS7Z>A)6jut;T6ZxXS*dXo{Bmx;)m(i}P<>?+SjaB6zrA7ODfE5LdNFO3fy_EX!c zspq62x-o$CgkO;w5(d80h_2Y$JO>}@u8ens>U{AYg~WLOFw_(>E~D)IDhgYGV?pRxI;o!NEXzbry{ z-PmCmwS4R-481g#gKa9jHdS~yE%zzAlL|GYco9CSJ6PGs&merJNADF9n6m-B+HTJd z`p*YstYQF?7b&@$e%`|N!Dt4TelYuCBfAbDJ{Vi*_Q{9&k|N)rvb51 z$e@oU;XV^L6A0(PRiya-KgV=jr!GP^l{o#~U7zlJ_>O_x#Js=^`dKc4`d+RV3VT?A zS<`$AwKM(?iZBMbXnj)W`rpC998f zZJ-hd1de6S8$@^%Y@@D*+p&Xr@qZ^<+8@_>B_0) zTMN=8R7E4ZVs+i=5Qa;1%(OS2=wvhVm;ex#uE2f4hm2#L74@%j`-rH_Uvj-AWmwQG zp-GEDJNEcCv2#-xi}IV#QAAd?H$SNQ+pWWnBswDgDMX4DUNedyy&SH~DxbcXNE=iZ z!dPm!CB}SS8;OVyYn|adf-A&?^2ZTTxsD)$-S3x!(?lCUiSNUMmoP&adK{I~ z4(3}Ac|(2BDrUJe2o=T57t1sR*}A=MGd7{*)-v3~p+VO=L>kKp96iy-eJD|(x=Iey zNkJ8Bxz(EpI-OyC%9bZhA&h7pR_U!_?5p}pd200D>#Ro@^KP7pZbeUOzMaKlX#Et! zwd5=8VP%~YTPX0Fq^Jaom46Nort(Fyc3WTTz(Ga|$WIR~e9pAO={m~_04G4$ztT_6 z>H4h;vssG)KLL|6mu9}gxA1oN5b}|?f`Vvzh_~QTrx3d->JsvA4JdkU&R>_~NXCiXRi3gL!3;Y0Wed4ZL@=K$B!(FFWt-Q0?xrvYG1^dZW(@vw zP;hE!P8~QlZW!nps?QgcyeZU z>zC;^OGZO&AsBSUC6!^yMosOu&W(yMR!-)?I@=Vi4eOjxDa|(t?h$+`Bk(HU0&fZ=ki&>Yzz zKvU9!aWkhnvCxN=seZ$NqiIG_4V-IrM`!AzMS#4Looa_`T@3@^Ef5+2VcpkTw$9$h zS?r?1A{ijj*4zxE>1uhI`Jq}wO`tsj+B~cbs+j*~0IPsT$m{$9I%iS*_EjZUDUFyD zDN-RD88r=Wk@(}5RzP*J%0!Z7P*)4pdt6f%PG#*z!TXa{9mH3Oy*4@{tJ2_DEH6?F z;1xIa*R@egg!1B@)ali9M4B#F{}rJ-^IohG?U=RxAd zOW{QZRzHTi^Lb3L)VW+|^pVlYBQii#iYX%H68@^QArcN-#Q7)_oksp$cdYm0cfGma z_x~R~^pWOh*J&{WtdxsiBztK~ZS%G$k|ftYP^$bzxCjB1czN;_!#LwE?PLW3}7@a?_C%S^p>v10|tV6~!*3A+imNanxRHK(rH$x6{ zR&;+`lXNNH{{?!y6~-)E$CI4*8lup63h|u%<9>OX`OK1|8$+XeJIiZLGlh+82LhGj zg>`_c&sZESh8J;9C}NS$@~+*hQ5a+itxx=gL|r8AyvI#K4R$RX6Jxd8NuF8>)(2Up zD^U1|oN8&%u<}S=2Zp|uhPfGh-i2(o22`IKkMreWw*zBqnzUT-b=F1 z>bNs$KX#3GaHY9o)=v!&@*ZmW{DY;9Wv-ej%1ob+79t~hj1I|8#;0C^y0%Wt_^SHD zHUu^)k1bwe<#iY)Y8eKYuCgoch-2I$@m_eXwza%!ONm`Qy{K<~#-xnxq^UkT8b3hb zonZAtu_8%E0nz;v>+VHmu zh~%wjvWR9gtlT+ooU5Truz&d$Ij#fn=zGx&uA?}Gl9y}KAN$TYN?lxnCBZohk8c0( zsCG}EVL?`}434Z5oT=+Ubf1BNvU0PrJpr_)U2CrJ2j(78Tlqx4Yq1?tj*)cno*1L_ zMu&SLgP4)I`@>KJ7tpa`g-HIh zzh5q(S}WGB@3bNkSX^$o@ow1>Zu^&TfI_`b!OOM%Lu*JisgL>b@qozg18X_MK^8l8 zamN;!Uonu^Tf|GDG?)_OmfUV%E_4Jtr4CLSPXEzw3Sb*oH11zA6LzZ#+`nYYzm_^$ zNf!D@Z6)DutVtOUVI`EHEc!+XPRfU*Q!fkOzq!}eTg@qc7*AuFx(CShw4SwK4t%V} zHcy9k%?>$K^c@IYTstp62kmsCMBuaXGyEJmC4|1aZ)Ic=y9v$lqpR%H zDq_b}-J@0H=h&XW@Zal3;X7Tf_>)6ROgQxLGHJwl{v)YShh>r<8ET)l~0$5KVBJ0=D=weJ}HmF>bWNi zToXm#IcG{%xYrAPrF>Vlhc^(IkNZ9sk8QlU2z74G2g+ccBP+B{k(M<@{Xc~j@>K)0 zjKw-Xkkj6^h#sooC5xTM^=k@2=iDM9wQK^U@Jv@&y#eD{AoyaC-}HLy3viz5szet@ zBeE6l4&4v)zNOgx7d6La$}#+0Vn%*rUD;jJ{ybjkVRjm{?y56}y>h>8*IvY9q{?p| zP~7apUCIotC4LU9B4!^0ZhqXD4y!C?yRUsL=6$w1dDuXHVh(Mru-3|;`V*^615t9t zhN^R4_b;2?;IcWriq=F(k}9!p)}FtlKO6GwZVCNlAn3ioRt2!c=E((DcVBsXD+(Y{ z-nk0=>B!uJH=%cZY2ZH?xW^$d1!lEK354D{7>&*3hEWtwI#XcUY?>bNLkel~1pD() z9f0(UaU@auH{@;ysH^0ME21NP|QsG)}6z__4=0!z>lemzJBxEg!Ss18X9j- z4vvIS0qNXLk3WzvSk?o*b6H1wCO>$$yGp$W0ZCRdf)HZ==TP0Y-l@f*eSE4$y+m6% zPI4;3-ikR^4&Gs5(yJ_qbq^MY+7d;2$gRVBDyj3Ezx;C1=(!yb)Cq?%`P)t{VqO52FwK>~BOT z6n))^3|{AqP2R`uw>9BPA?dl2}5+f-3wM zE(O7b;q1pqJBPcVd%3z-FOlLZcvgvP`;p2Pr=anUvph&FfjD|h44{SIuZio;v%a@# zcYKbG+mVkVL3ugDX|;jR2SoBeJ|~&T|6=->W9f_&OH(SN6l0r|GPUW$y?KTHRza}ZGZ_N%>}NnySYQbVuo6Oj-RTz>R2N0I)?T@LV``sPKfFr ziK6Jlif2AN!XmKD%0B`{kq?`X&Nbp)C@J zVii~)?GE^zr_uB4o&gAtYS*)=**k<4Qe`L3H6XG6u-3*qea+{3RLK_g8a@)n!4dde zE(M49MXFI%!j{sdeCPteS_zVgBIln-X)3F-P=-{4%zuKD)lEOPSMc;>5vfFRgXD;N z*U^%Zv^JLr_O{Z^bQJ-|7y8tUcdCY=htno2>v#f}=!4jHnWk4xL1v(+v_KfE$@kSO zUMjJ)17hyQ2Wx}^CVL-XIK_#l=M9v)0nWctOI<;gAsIIiRbj5RZ{gb6W3|hf&$lA{ zIIi~DIo!`O$Zcf0-KfK$zi#%yp%3oQ0T&e85AU2{%bgA#P7inlPCCqvO&2h9+9p%_ zepww{E2U0b@~nlrb}ROPrhDaN8biob>gQkwi^Ve0jz;H>hn&KI3W*9Ca$C$DEk_A` z%+$MMfK~(m_QG+4L}zE z=u0nEFe(~kGiL+Qrc&q4zW_kil16F_*u+U=5;68!D3GlKf6<;&ay8i5wjF1)n0nYm z3$_)a7|l>#mcUbHT~tgG`G%2>l?D!2-tMSsB)rm72bD9=mZkV>xHg= zOr}O*7thOIBt>*~Dwl>jWSWe&14Y%=-165F^+c%W_;(!U$EjUq*$%o0o==vasR|uI zaltr^r7vg?i-XagtT3*AF1W}C$gkFxV=|WE*FF59A@O4S-c64?j2Dw`bAOUryI&Nu z`4Qn&=7d9<5;s?xmxB7vUK44)gPCPlT+i9sVi0kXNd!02A`iSqLmlwaS-OlhDs81l z@Jd|6-4V?vtQS(q$xT=67b8j>ShLQ==UfgwX4?NSr{24T6QRxMNCcMQ)nN}$WaCtt z>;UGH2enD`0i}_ao_M+&8vs=TrwlU)XciLtSpOi%Dt2O-cW?8eOOCOb0*Bm{{2OJG zTa=v~ZC4t)*>C+dmgX<*665AhkFrti)712Lo}bl};%a?B17OAww;AJx*(U*kOO)X+ z@mwAyqg7>mF{&O*^ymWPoy%lJ(cYtzAEp^K3&}9&E$u`4kXJ=)0PCT+8S@BvI)K&z zkcZ*Hgw3qgwKHFSkB(5i4O8As6D9zNM_q{5F6<6Gk!o7LpmCz${lQvoDb;vXf^v9X zo@xxMqbg^`B^s{P^Y5pVEe86B4-smtyKZDJ(6~O#B%y19+5==$^5|kKk zo~L30({qkm)ga;)+sHGKkq$}u2Q&>qXIhC+J|Psk;*t#Lgv!;U`fuK#d*|Wg6DfL! znG%*eyk_}!7*5>FRYCsKl%-u-t~xjUy~7}lh_;Yc^Y)BF@OvWl@l(E)H?9;6_449Y zh}lflqV3FtJPJBmAl?u}k^Eq5DxSR|UX=Ez7x--NZtHgz6Yg?mu*7 z>)JhV>&-i09i(5Hm7LS`3TwcZNf#wmw`LS1QE~%sOT`!;WSFbCjQx515#&{vJ+P;@0e|0SDNE}>_U68yZ--~;9O$**M_|{YMsFSmz)gy!%qvTk zsPo+}s?~M<=b#ltY-xNLA79F5&rQnSv<_j$AMw0X9Rs&nuh2o(K|!oDEMyCRH1zg- zr{NFfT+zQ}xswH$usiR0z5Fn?8{yB-Vq!E+-jv~p%D}qiD$qVq=kCY1i)(@?L8Jss zp=Xq-O9t-fuySE_(BIUCCuyZz@w}EoOm8GQIbI7% zzfE}~Y2W)z)#Xs$W60+9w}%&Jp86>-q2cb+r(K zA09*B7T!5oYNw!!L@OsO5{SA_|kFu;h@Qg$SbhXXgaB;o<@$5spQ^ww=Il zp1xK?5kDufix8cjsN@cyL!GyAgZLLhc3)Qj`?1j+UhLAO?F^nKG>TiSsWtZ*!cd7D zeW~et!i)tCf&3cpI3trkBt^?#mR%Zhwz7^;w*AjOu$Ot7fW*a$n})ew;vy;paonJ+ zJ7f}ah`?a5w&WhtUFb{$J-WZ-96a;%iA#8+O&nh>Xj;SH?*XD|4!VM6k^15pgFW%M z!cRVMGYsRF9MM>-7W4uS3Mmu?RsdhmMI;dzSYKI>)>zF8JMwGoGqCpS%SYib(QR0= zsKHB~Al0=eOaNKUAT~u_|HhfC{7j*Xf5k~XSd^)|48k)oBS%cYnxaR!n|obZp4^(Y z%k5N|6RAsw_w|W2nkUgKYo{;biecw0Y4Am*#pN+{sAX?;l*=&_*tC~6a~gU7t>4cy zG`y5HsT_MrRNlCHl;)+uK1SD9<1hL;Gdc&m5u6Iy5*9&5K-zShtnt zJZHbVKtr_(1*97|*4#Ejv!liho_(=Ibz-mg#dZ{1{d~toIh?|R2TrMR{et|`y~F91 z6%0&mM)o4RR8uOkg%cWs3Wg9H`^RyQpkeR{;QkM!M(-vm4zVJ7b+VDlf+(;6kg{hV zS-#w#CiIdxjn!JT*F3@|LDaQB02m)Cj_jgm+}h3F#&S-EMPG0~wt3ehB>-C87AtoZbNUfAu}V7LWtosR62{t3lNxwd@1t!JyfOD1M*Ywlsqu^?S%}S&=b@QeyXlYcB=Vd@)V>F!5dc5dt8&h(1_nus6mwes}*Z$AeR{J)6~^dhu@Gj7u+*WW>bu2%uvU68EX1IB{;>Wgq%U( zIpWWHwrkC60NIrVJEGQCU>t>L8Bm|07g(2zx&0Dq#1t^mrW|RXjM9C`X;r)uyj7 z7!?hp3-*C3D12v7BobG8&b-I1%gsH^I<+)td6)$;D}c*SCX%hXFS7$9ynyuN-Sl@F zZsudgl&1=@z@V2fFLmD4%m4`ouWMZ~mqj64F6+j4(3Wt%v8O;Yy4kP3_Ulk>UaH-Y zHz)VM2xSlPZN1!r5~l>W5GiC027?1OIVw`Ki0)X{4^#%Jj#AAzMBmh!ew z#^ODYfcbYl0_o{`&4ErpHjEOU7r0*6FcCR2c+2d*E0{C6I91Hr^K9pwrs9Z5zN|FLi?l}k!n|{B=0zv1)SD_dXsaDV!LsKpoHy zC3k3TlAC{QTM#pYAUdfe^6s^~wPHS(S0Tj>;`L1_&?lWj%xCcaU;WQ$U#txlW9EZj zRE1!23fImz-DawT(eTUd-X9ih=Gu;z$v!i!Sfc^hkK9gQ^XmJKii1xzg843 zev+z89yK528xLCW_q%jM6W2i5rzt5*3+Bw2vfV=npPI9?kQA}p3)*T*$3}ajDU@3t z5Dgd)p_(E1?_=;i0A+V!P=yS#TdBRiLflAX71FiNh41sU^w@2`s>ot7eHTvq zHz_m=M<&K8077@BExck~<;t%=o{0z|!6HFlYQJQEop==F49<{A>Dl3NPtkQxiW?2QN-=~NwGFR{Pa^>=1A@`X6 z8g;RHqeE4ISsDgXZwa2W=*=%`Q5qA`Hfi&#z$yWiQl$0IZ&etr;+BGS?|1DvzVkMH*rsD37ZVs1+~n+N&Q#fO-mclq1>uh z7Jvzv^1FrHNni;fQK^GDw!QN>b0F0eWmZ}r?dtErDXDF18d1^;Mh{~FnL^hbgY%PV z(zMieO|A)019hj6Z>k!Pm;G7J@Z}YprZ!mr;lA!Pw~Oar&+NJVN(TE?g_D)SRTa*O zFry`=aYbpu(A1jMF7dkEzbW_BuZPvT@GBD7jwDdRipMRLNyQS|;>(bxP$DU+qCrho z3O_Jw++iz?o;*70-oX=ET45;%_UiV!_Z2Im0@AIBnw;C*1^4DIxt1VxlLvjm3ePrV zw-{F1DO5EV{BC_uk5F4`P`y^milYH3w^u0niZbw6sv|?F74h_e>)}X=*xr)!__Z9L zRqKng66aXm7hs{)23BM!bwAd@$Y}BVrlF$TvMJfgK74?rqu-DKp%ihd-be4Y2e&EV zGLeR0*Vt(5%E>a#VegLpW1F8{6?Ci8O6+PRsaCQZMgy?}qti3!fF|&7qI{8PCc{#A=b~T^`dG zlR0v{^HPl5o^F@<7kk=+2XVssu+rTmX;G=^k8NG90+7oK$ML(VZ|0QXu9aaUf6{Ah zS5S@zO&La!ISS7aOwI6VxnwB;pq2i{oLJR7CJ$dI8Ki|CIXf{RL2@dkKSPd)@;2dC z=Y};GGo5zllQ()h_|LSnm1i$9XYg1Mia(i=^RvI&EDxQE_V zN&{#O!<9)z6n~Y2pes27AHSw&r;97Usb_b@M>s68`I`?BJX~jZ;x4(4Ib0s#6vH6- zoT|1rEo6Q{atr*2tLHEXn`KpxMMhjSDUf&P@PPdC@>z zcAZg%j2D1#+87Ee$wq}e-%%JJR~3-D4I5>ueUbZ1;m{ZFYEZ#D8}RsQ z4Op;h#+o9GwZux8(Cd(lhNKCIEaS8qm`B{RM~*#4*d7wH#fVdQ#g?EHQB$a_tti0!ESW?}i%v|goa zry$)YZ`DJwTneTGi&-QNrXl-rAg0+Lk11nfR(!ny&P{^R1HYn*e}C0IHDX(8t%$cD zzHTZ=tJau8|3uI$ZOjRSYGWUq(_@)Y55-?KAlx_1kny$v$<;aZ`)l{fcjLDphP-xcxGCHm@zP` zo^Nq2v;bkgPZ@g<)@6kSNea&mnw-IxX86$W12;4b1oBvK>2&aj8$StIT9DP~F ze#0weQtI^joogL4J{`$4tEMIon%Qgq2Ilj^YUmLxT5* z($+&jM^oPF52#J7f86>e!;F1@aW?4G^7O2y_fX= zR!Fi}6oRjAlv5Gz-bMqt0@@a>^z7kteKTJHqTR3__1YCXusnoVD7SKM{A?nux5LyY z18D;px=HyHxHJ+INi+u|tWY?4^4R~;4UC$)T|DaJacmPbv$RYJ;~=H`kTM@Pq#elm zxe--o3zsad#8)cYjhvvei<*#UN?<*It(qG6if3N4Zpj>nr(1UQdW#4HwHIOn5AGaV z0*o=No!;3kH5?Shs^+bYw_o3HLCNEu!B|Wqg0ori93bDFHb*aMeO;=BnzfC7n?q4E zRZFvM+xWK_+q#hJ1}wc4$9(^ebq4f0=+?i?G{km-O|huIY|W z7l{-j&;#G9G5L?poRzt zL2V%27;Kr1`KfGqZkq7v{ygZ1zoTtVU({NIUGop7Ig)O$ECI_K6T)QP8^&)@VWTe& zK#(GAsVzG=0E8`^mLtb-Bd&=oHw^Y7p>JquP7-CZZYa! z0G>Awb`w&pvDs^f3V$hR=AXdKC%fPlM+yS}mFJPy#Ret}*2cNpTQmC%hHL!5IoZ|< zuIJlVjb7>Q%Wx5)a033c%kq1r>Y)t&6edRVF2qm0O#*;in$sEcOxSrPIgj)iI%Y>* zO|91R2aJhXEI<6C4IY}i$k9c9CQ`y7g=m+t&IansuKb?HY!@!O$O$1J07Y7zQrQt|vI#nwssYxjL5Y&}hToS%oKc%H1;^|f46`mv(?Q?*3@Cyosnr7U{%>Kh!Oy~7mZN;_!V zqDQo|eAEZ2Bf?W%2NXMWn^KsG5VC>c7L&U8m^S+m@j+j-=4LfvE$gdmn&YPpI+hqkM;=05W4#NS>txSQktUrq z7K=Q;*l+VQTnYD`mg1~-oa|LP+@zUGTpnrwJB;z8vH5&k#ef`Lum_ddq*3uJ_(^d7 z^Gzd^g!YEwA+wDa_rYuoc?dq~KF38LKhQH&Qk`5`^inD(eRES$ zsAT&TQv$P9U0U-1<#Fh$ce)#?L@#@(`nqAQM$^zobMtRD^dkWOK7d|SDyNS%`K#BZ zf{ImKU?p^Mg@BNxJnrRSSHcmaN_~B8h~)Ao3(7V+iGotc`oK~AwQlic6PIrr;rODS zv;?$#Kc;MBdIdhsTEfsy{^##M~a$9^64QvK{3gV zPRXNW{XAM9{vPAIK1)>sOtqk@TZ#C?zeS;OCVl_HrH_8@JEKQGYLrL65z{wb-a)x* z!1;$K(Cy;0IYZM(#^56Ggbe2CHJSIId{rsSMkE9}*B>g_Jp@agA}1#kByH<;>H#U< z75XRse>Icm@$Dqf0=%eIG6D;qgOa1KKMsd3o%85fo(Gr_wvd6AKR(4Z!^+YJ$-1Q~l z>{~@s|K@0%+nKYNQCDij)7BUa0OBVgeEO(V6fe7gV=;q;A5bClqIxC7Gm zMRbk$?$b7P%9&kN>HgY`#Ik5Hr)0?{lrM1z$4jJl!^=zwoO+lPZjOc{EW-Sd2w6)} z*=iBz%%XWw%c-a;?@u9E8a*14Xx{$c6Q#EQU5H{EWszI{&!WwR3sbL2=WbwpF_B=! z&^u&vb2#ncm8sjpa?LpT6Nz-=w>wJ^j_5-bHXoEBzC}rbY}>Jf#WT$&<0`E%QC%AE z63$5f&Yu=P2CZndK|ag3`U=c_c4LbOMbm&AOD826eb_b4ogbM@P>7ka56wYx`q}(I zB59s-v<;e1V+mI{2;nfLQHM=PL9)7&VQofdF$H}lfn1o#A$dHdb?!$EAxso{Nmb+A zU&C7<$P~k*36~!SV0kaFo~WRVQdm&-5@9OR&?M{}M$#~nvEt4{R3XS#mYWIC zr|?HfY)r0c`&l~i5bX+lno@)~aCS7U-RU&}1Q?{wY@8fxTAuvG#7Qyy&xWtRj9v@k zF$KlZE18wDc?P)1k0DKEa}bXMMz6KEz(?H+;_-vLICnMu2ckN&(K(8FDZd_)kM6K? z$)SNH>8-|3YAw7kWtzE?2WKo5L`CXxzLKJ;mn;;!Xh!w-3Q|(U_x;Ypt|uG}K4rdE zc-oE0U6dtbakp$GDC30qiMB;`TzSlVR%x+z?c&y{!v735;hpnhtJ`2319zX6hndmk zmJCC6?-f#lpPP?Llkq_+Gq@vkB16~IRl5o`P$kVLBp{1FHa*iW{~F2}HGmwDhRLu> z&^k^44*UVEqFQ)kU>G^v%(8p=fUyJ*xqDJVpvduRkdo&~+iq8V5rjjNXutt3n5c{4 zeIplozONHrAis@LxjyFf)$GC&E=_ekTA#~Cm7T4vA^e@}~E6G}ZZuDqyO|BNd8 z)W_$XC9Le6yP`?pd!9na{8o>>QEnoAo0ik{PPbuygb+a`*s9d$vBC|{G-G~Q?M8M&yNf8!H3bG;JhxH&w6B8JR}U(Xk`G#=A?kG ziRvjMs@_!_Eq(m%@Mo%&gzmtyyGzXTC3TH7%cQ~9d(;*_syCQny9=%!P(qXq--W$F zof8D&-U4eAPR1o@Rnc1!AY8LXgohLJd72;z_pGXL_GgxI`2foIV&!v#W;S|WCc!y3 zo|vhNryq$WzU_}b9PR@=Ct+^2oMSEP27NhB|GQAms8%mh&{MAa;Ay#@T^*~sVrk^;aOmvklg@l$5ZY?2Ye3gLk;IoL?uA_h;s&zz`VfBg#6 zoN!IB6_GV!lqDsxZr0%SKHL%O1mN-mn{^q2Ro-isDSTlkaJ?S15c-{@HJarCJNWYQ z->m|+lOEfgc!dHrVlRG0c=Mwk;h`=-<7`*RrTR?^9q-Yq6YkifL_{TRT@nI2@o?R5 zDYBOX-Kn0~x7r3|O`72+8;gM_^#fbW@?<%NMEQmgc%WZ-B3TxgO=5zbo{{$CClBUz z>AfSMDgQ6Gfz)Ul2TIXJJ?Sx9yJ$4>*ee!@N!eYTFftyEkY!im$w*+9T~&!UgKHP@ zxb!-u^Z0ZIY@V01d(x}W(FOs+{TXeexyTojHqApO5bSgDx*boNGu&C*&{oK%sxU} zx3X~yw?%%*G0aXCYz*S5D)>~HLS+7YqWRxL0_AKkOzY=%JURexzRebop4s;j3f&^L z+u-o12tp7<=hWl@&p)4lOm!a!YW!WQZ05;_sJHOq(m@mCVxFs`#08S0VeKj4{i$XCXwKg#{wBTxgIS1MJvkMa5 z>v0`dUPby{mp?Pcr>rwl7ZyMi%@Qq(7@?r32WGhS&=S~?JN6t zH3jd@*sV*eL8?wpXSTT7qE$y97esTS@8LI4S;5V^Li6g1Z>yhy*?%%n=RJvU$2)U- zGxxljlc@G`kPAMCBarzST^p4ONSLLlh@xW*MT;!y^q24xnOyVm0N3km5lP0Y!@CJ#UWd89AQ?> z^Z)2vqGXebr}39Typp~Gd5GQDdF`gABnYy6z}qwVLv_9cqGEn4M2T0tz)_!l&H`4* zy>I-aFLzG!F<6VV()v+;h;-@OaNE+NKKSFoF<`apU%2fPr?;J~vz|a?AgVFec0n@p zM(=RXq%T(=ZTjSIML1`sqw~YCTzLDdHUJ?Q2d&7QJ1iROAI2MmJm_w)H1EobG`P8e zjfo^{R{ynp!I1@fUBK-XM{en>u3mI)HR+SsTW`Y9i+7gYYwD%1sw7HyFNQ|AoB2gU z&Tj19O~2ij-lYb+Qu~vTHSqx+ZHaH%Ws8Ranbz6bN-6 zu=ZM2MJ^4NqE-@HkT67;Qb?tcI2%-$(cX9fI zGn!rl3il)6taFz*KKJ|_*4+W(l;y3R>n^{Hc9vwn=oIv*SFOP0oQx8<=fb{no{Me< zIq5qYVQ*`E;@k__gnpp^%ydmx@vC_fa#1nW>Lt1C41k+VFSOs2yNqX!>C*QeaXJF8 z3}dZ4r#(RHo*@-O>q-}+&%QktEV&?@66@l zJtCNG=Qpdd-#g8$y9!Qheub~F9v;q88-^j7<=;sSJZLEnSa=e2Uis2@waYt6vUEeH zMU|%YkvI_EvJhy}$8jURut0IF5QBjNkn1IvXTLtIvFymsFWV-2w|tBj^_anShoSp8 z(B@%J!N4{1CaOaE!I3*jPJ6)HQ)42Gm z837>*_IpuSaP@E6il{kA830{*^Z=wxS~!+8cv$+Gq{#0)#T9f6=-V3#6Sc7%P!l?`i6X^CmVI!E%hDVc{DG&etfZ=Mtm$cZQ&Nht0p3;~()y&NrMaTxW@~A7;OnPW@teUMUd~Y@=@`4(n(?yN@ z@Tdxa1#Q`VI&HjOmUJ;Iiu?dtj@R|B!AA01p>tN&ZusCC+Co zEFK5dB&7YyShku*mi;GZj~8tD<80P@XdPa7 zWaj?V%71LDVx*@47Z1hV*`c=ntjcr^#7slw$Fh^@?*DXbxCQrdS$3fChXe*XLDt+WjfspL zoYRGQcVOuh6d=mL84Ct8mw86Xmu5Sdl@1KeDOQi%>F`m7rOdD|@3gTME$(b-`sZp2-YF(gLXKH`w)9@w^MHkCA6H9 zDii#c;i61~xhP$Es+K*Qn8-8th6Z`As0fA>hV1a=g>d$6YaDd1hHNf{_S}s* z+D<(k`2Omu=Xv*rhGm!3TV=Q1uh?r&E$(Qhxij=&62{-LK-)kTPaRzj1+E@w_*2eU znR?$u0ny!=M9rCNxh~`*7IS$U0X4x71^c3_$mdmoGbBCy1`!6Y>aYIafrrf~MQ-ip z{Zbv{hEVa&xhNp&C0%EDC({XEZM>Utg5#V85CoVkr07?o-z-SCU zw^#a=JukR<`gBFtN}fa05-M{dbdXS1nc7uV#_>0kDtY=;Sn$k)hYdQ3UA;rxqU-pi zXfpMRbq}YeZPHhEaVS%@i2kuzxY4s!%(3h@JG4Ir9Ln0ng-kP1mw7!7k&RjQHmosI zFY&uqeB81I&DKW`9rrirBo$Y#weaLcDum3t;!D)||NeIx!i=A`cDt> z#lK|>|Dtw^bhSMfxH|&3bQvLs*(65B^KpP;)M!OqF)>;De+#h*0pPr}F6e1QCm?UkQrnU z8t4_sfUqYjXR=85@B!RM*mSSsj@BrlbX8|{hHb~)tquf55l?U!7kZ(v;;J7WZ;yQk z;fAMlz5Ei)C&7$LHFQ?Gy?v5thiNo@v7MmZcN21rN63KJP$~T@5YT`Sj13k-xW2!% zC>AhPSc03w;e}0w{@64-Yxia#ta;-X;pXAyy8LB7Qs=KU1tCaPx6 z!N^Kh235?^)#c&@A@wUeAx}(#jMu9ZQEP4w;i_Lr|vO?D->LS!Q+|oX;q4N053q$zteyxdVZ$C z!1R`~GKFh+5Miv4xJ-Orf=o8{pG}M>~ zWCQ4WN#f`Wi~y-_+%hgnV8r(dt2a-)?AKMKs}(QoHeSKhe?bLbg(mDZ<+(M&9d2l{ z`?GAvEOL(R9|31aY=MnILC`Uflyn5I=!=l3)mdU`dU~3F3vi@F>yEQ zMXF1CJ>g?~TvYQ17(}!P1USJBePQHD%5rkBBW!Am`ScSr`AvlPY$Dc*ajW1y#$zXa zkx2V99)t&b#Dd`U>ha+KQgv)+6Hk70ziIdllad447iT5r<`%3udG^re7yJt+DmAxq z=66stoE3TWeeMwGm85{Fj|y|&Y8lr@n06?xWY1;kM z05%dsNEEJ-gpvqt&F~OZ6m;W8y6)R+ED>IlP;$YUEZg}#6)Y%yo$oJU11`A? zv+Mr>%&i!xU&dddMtULMavey|W8+5ww4xf33dbn#H5vpmS4Q!bGDg0Ur};-z<_jJ7 z9TsL-HIpOF@8$B8F`2&m$d2!6|TS_QE%Edw?hH(YIDfJH|< zCXGjM3BTc8^H0JZ>+YD+iN)ikzj%&jaq>kQb@pBF9b+L-=M_Fm^w2W{R2;J;%HU+| z`sZsdAw)UoE;4bcbr8_xJ#=)mCV_r)z*RAOBe$jVu$14-3V5942kJ(pGJdEY0Et?T zd&5ulw0H%iOwJ*3EP@Y;l~22q>zq;6CfNEvyi`X2Pz7-P^j5eHmQnt}Y;bifwW0UZ z72{*<{S2p#n&^Zc>Wf}0@$#^LV`c(qiANm#+z#3q(`B?Evf`5ibeDbWJ`TkA)|iSd z+yW6n!VFO%%|{oXafYlXglV^^z<}>TERSurrttU!Li_Cz0x_7B%T{jF!(?(+)5p;l zZG*){_E$mH8!ge?q50Cy;8k0Ihgioe$uG;~5$4=f5@56YH|Hq0E5D?2J~DLQl{LnC z`6arnHfK8kzUDNlqfi@^CPty}P*pms3Rcy{kUwa+PeB$AbCjq!axqyDXL=^)@3Dfr^o zz<;2@>n_*W{Q|evktEkGgN6K!5)stR^`E=AP(U*EM#jWFni)R*T{hzJ*|zxdTCSyi2WmA}^4&cGcuWrrgXM(l0SYdyFPReX-_ItM5}3 zg%dtSG;Ayx8ar>(0sI*0Xjya6fm&Qr(XA%t3p8v|Tpwu_8YD(GC{@ANh_tTB@G?a* zR&?|)kIF0CEG*lgUVSL#IftkAoT$b_sGg0j9)2$lBwK^PphXG=o)Ojvly}&ey1CN> z@XS1Q5s#46>XAbOPnBO;hssR}Zn!43MgnS581Ykbm%n?uKe6e;It@DH)D&<^{F@qm z>0rxLr}rKu40S4jSs(s05QLaoEBC$12bwWo6sOV;J0TN`!w78WT8J#vWU-=#vw5oP z))17vHD1TPCQ4i2^z3&JK7zwSDs=niR(Y6=msp9n|ICjrzd27vv0Z=+LCba$0KMd*FK+H?ugT0wfmm* zYSFY-x7a6RtmIV6C)S1?YBhbH(bS-}Eqx#R`8RsGl=;gAsj5B!$#`An#_0Ux(AY5E z>68WkZvzre_FMx5uXV@Or8}7`wN*_UzIT3l>5TBwkoZ>6$HsZm1EpRhpvZr+qRmJV zBIx!v*0|}H=`}A;0;s{fBd;gag?~_*OkqFkGb>hx zG$(SBHZt5c4-cT;C%vt^J!?tIUO>kR!MAJVOy(2g2p%m&4-VJ5md zCqS6BN2X~SGU2AL%@M8+8LFrpt6rDo0WU|L8b}Az z5FGE3fO~%dNlp z{eJkArh%4PQ?_5IayB(=V$i(@PkjPt#!L}sVS4%Q6($37HH4BMBz2`PC6AZF3MU&( zf6Y4&mW?`E@2QR6E?Q=~)(2)u0NP_3c+9b>0Xs+$Kz>xP%D4gG_ah>2oskM4qNVk` z^w6B8Y|%Ka-u%8TJw{-Jfsjgh?Jp>I)Sigts z?{_@Y8M~&!$qCO<;Z%`Z1Wfm;yJY5lI~cj*SwJQL2RThV75TJtRvyNr05r!y)3(c9 zP31JeTg%eTVD_Y+6Q`cKOO*aA8P(YFmHsABPRRBWgZOUZu^oRM9el_Kt{wGy7|bO1 z4ikDUS4IRRW4P&1w^JNyYS>z1CmuZ1B+FBB%|2ol;;%QZ#_>01AL0e~6fQu-k!&78 znh+km^*+T^7&_gN`{b(0EUP8n$9d23AzESZ0hLpZM5BTrmYT|60?W7;Ju=gZF1PW| zp8zj_DCReH(rm=u|EWKs+@jweF|5+GrNG%Wnq-8!^nHE932c~>UC9{G{x^^e#=xu& z(Ib?zHbA<{9;!TIdTt@T!Pr|&(W}M@;@!s^(-SnY`ajH)>CZ|Iat2<(JJXT0d#U4a0~<@QXA$%+&PhXAUIyr5 zLn8n^LN}(XTUjA;M6P>k2TI2tY5Fxfh=vHc9|(NnZ(i{C*3tp8f!2PK3b~aZ(0KvF zxsS=Ve&;vfhDRko&>R?=8pY|#20B}vaj^c=o>NuFNHYLSW943AYYr9{{_r4->QOY6 zi|7x1aZJ#Qe2JT)s>I}iq*D=gfjk8B<@_&w6e&_fphdef7bck)0 zO;EwzL+{Mox`2qGho0@C&uQaadm30p@eICq7zn>)>#LwYqXASudrRm$L`^gfR3ieR zf+7}GgWV)f9--=KgShTS=}=>IbSBjm@2GJf6QnH!N3|;Ch&(1Lf>@bf_DMi@fB!vd z-D<+(s?8oo1d;7647GanTj;CZ^F&1!l{-J_vAZs^Ty~yey`w$Zo4(BCwFIHBUk$Ie z!VGjTvB*>fR0O0~@3Gt>TYq^7H-EcAe1beKegoyIXXgWRvAG9Mp?T}(7^m~V@l^!q z?^{9Ns&NxogGD<3j)yh+^uGM;wNqq$a@B03SbWzc96kizj)3g4cBwBpoD$!x4^n{_ z-F2Jv?u<@9W?J(+av{e`d5$cbca2#u>ZjPtqv{$G82aEsLq>uE!0``r^8C+NZGB21 zyyLMkmD&1~>Z#86@#9KOusEKG{d#}gEx6+Ep$?x#q*Wix?sq+xOxTTCk{z~x6Q8F< z2V?1k@hF|8tBf%+J-KHq2hgVOkDnvP9X9-Z{rR4N-mGL-!3?gTE;8&r*Y2Z3845PS zlqpG}RRO5j0_{K2*=4Ak50Th%_3Hjv9E{sMg=;H9HsGc`Q`h}Jga8pSM7Wqw;@+GM z-`_!@>7EkQojut*$00F`kAfG3MK*R$Kh;NF4hLWT$Pz<^P1+*Tr|EKFBq zeK+#-u3uMoi2v8eC{{=5W2-6k(kDKOT4hA-mR|$uTdgv{SP@8XJhiYlHd?t$sVa=i z+#42F{IA6)A~LL~r(}yWNG8-kG(CfEemq8Khohddk^yQLm@kuRZixi*!}h0V>GgHQ}6B*YGfQCMz-SrnSQ?jrI6E5?DoN@>y-=j%X1&WXE37 zg{3I2EvWhfWrQiTslx_LD+cm+x`Za{XB(-TjkO&fPo{1J-L@o%r?|a5a zG>$SpQ01yz*Z8Q{egs1gh*r)>dmVQV#haO@H7y?BXj|fZH7pWj{jl_1_P_fiB{M>0 z3GYs^@G&1$m?xhRC8TF6XvoyaYQ+SSX|5dEul%Vq=MPREgNM3$&t2%+xK4)S0yBSV z_UrTfWOA>);>vX@2^jk~)I|ueP5J(sB6*_^8@pP!AU<=wZon^}7|Bo2T(=jr5qt&g zuX(c6Nq(JNYim!>eik5yG(NB^kE{WQb}`1imRNjk_Snv5EXe7jAw}|^r_Bq4#T_@F znkEy+z1!#hSD6i)?XinQ{b&kzfVHIT-+?-)HhR{gHL^_LvSC>j4Rji6?2)1p+ahvG zt7zB6(YZeE)b6Y6CTo(2@=pZ$rQZv;-ypCEj7&v_0uWd7QSir{jI|HZr+3(&$g5;_ znuD~jc-uBY4@s>~!QoL}(q}WiOwI9WL*6a@=UZ(tr2a^@s2@QocSNJ_(QO)%yJS=H zaRg8+M{iZk5VaQoO97#m(7k$wP6(L&&W)AojqeOa86M5Gvu+wp`f_wL&?VZe3S;8) zmvrQpyxvy7gACx3bpo7XC26Ez7^p7sg$9hVV0Gm0Luy8}AEVDiczJR&GnwJZJT{)` zeADV|3-m^6nRK9Vpb2_@fsp9;R&3Nr0WdkrtuQsaqly5U!mmy*vP0yRi*oEU4DJZ~`0 zc-aU_>e7v^LU(v5%U668Y4%p!B8p>TxR5`%RDWz_NidGfW_&?OS@9@7k4Rfm@NbG} zWnPvm4f&P!!Z}a?lc4FXpX|5kb#SDOhQQYwKj|iwe@OA$;f1+cQ`h^=m%1mU?>OY> zoF>PxX5blB15qevr2Y^0Pwq=Uj*&BSQmXLeaoEm36)!r5`#aL9 zVt;WmYZmpeX5;0FhxxaqyOFMA)O~W5e{!v@kgQ@!I{7+WMq?PLs&O4FCo5$zj+3SU zOGVXpx5bE;SsmLDuY&*6LRwXQa$b@d1^sgKuLbz<*DA5xI1I9Z7-A=P@l2D^=0DPV~DU9sbliev+)iEiX`4_{(5m{?g!=UsAwb4XlP6E-~Kf`@9x zbBo)*KmLlO#CS+^Z%xbXY^=4Ku|Lo%nfJ}yMhwomcIcuLsF(nm@$NbmcHDNiWxSRv zbZ$0_hn6DZNc|SZY0ISqFUJPF1NUH`qQ;Kc15MHFJ#4FiS8}G*PSLZSEb98;=N?IE z)|@GSQh*Zk$1Y7TPG3mKmW&5}gDXRRX}SoQc)J~lxn5TxKz<)GmcJ?iMwGAKj0a$f zuIUS6^R#1l54HK6G%jBGZ*g|*t)*%O{qn6%a26U}S9;kQ9f__vj&^;a&de#45R8db z%l~D@>Lof^fAdyyg@%RD+RQGFI`dAT)VPd)&VX#g6ovW}OijX`(X~R|tR!He5z|JK zuyssJ04TVBzQwLfqQ=;8V9keR=EUwfKH9?)wz7?4!Bl9d5HYhyAuH@1oeBX){gU$E zdV}KdR^oqnibiHfd>pni#9Ev=JE%n!LXT<&-S>-f))yOci5v`ubGwPRT!$LfPWVaX zoW=Te&L_s9rnj@qNp18N_qzI5aj7xDyxWALsVF&9itM_C;swCn=4PPcwEmAXyS58A z@Y`Ae&ny_g;dz(T{JsP#)pi7x{t1P>@x0r%teFcCl*h{?Hpuw+)QC>^>7kNJweY3I zG_{15@Nr{DWi!b{yC?SO`*tp@L>=}b-mRKzy#VB;zZ~v~o(MsSAKq%}%{0XX(&B4m zEAQ13fNuK~Xxm|0wQK&RY-v&v6cAyP+0Jxwhic9IEG8NGCuRI4v!eFOO)99Z*0|am zD63J2(|Bg6Z`7Lv51yya?SW(O--0ww7+)iMCeSmu#%r`$R9xaiJs|{-ZH>71@nVA1 zx1mD3HO`q%8!=XtF&&zb-0KPI=63YRH!0I-l~yQ&AvALXUu-~xGNC5IcVGVo~;!3bo1YjPuO0Bhf;!^hpQO%+9!Rqe$})x z;K80B4qyZEkrm5tEig3B1by`ig+vTdfm8)MV;jCJrFOp(0)*(9M0j~Yr zS%-fj1C7tqxBR_k>v4~I|JTBTvdV5ub+>M;X6vgNd5jvHF};9botp0j%1CzeKYnSG z>VmYx=@c~x1yg=F{P)LCA!t?gO4S})sp%O@auD|cEtx^IkrxCxltzG-Y{=6vu0yfR zn`FPPA%4dQXhm7!fLF60go*ym(6cN)*^Z5qx`6&q`8`{V#lbE~f`a(j5@ z;R~iQ!f79Ukj4mL?(lvRaj8VVX%E;-7Xt1Pb9lf%%-SU_Kc1I$^=9|wj!U)=JJqBH zsZYEn>cr$SeM|w9T+U$pEv`gxWB+{ukr2#flHis_vix?;ZF9D(N$3@i(pYxdj9W~K z*>vQ00|lBNiMmNgeizWSNYbp1figOPFuntG%oHquAA1=9ml)NfuX0y^=;y}JO(c<` zYU2$}UT`Qd3aHk^)#izvQ$o73e+ECE4?G8pYm^rS5?4mZ>>Yj?8n^*SaM83_6C(${ ztH6fye8cFc&ZqH=3w`?tWUh1iJ+~?Uk$Q<6pM@l#gdBpYkt6B{GF{~*9_BL27-;H< zBb&Om{ncSs#Mh~~CeCXSs40K#;k$a~-nf9FmDZXB+dP@{?tM_QCm~{XULpYaL&1t7 zBxJj>ejrN-5m^zqcW5GPN0-#EAZoK1q@JrqNLv3co^se~8QZ%a7yIL%;-}oX$%w_h zp}?u7iLD*Z-pxvfGBsTCmG}7NoIA@C=!UHYmDWNa6BBal*?p?Upz+}f8cUglT5^M0 zL6|N&uFr1#(D#(95az$AdXS-Wxz4}cF%5IQd^ldQ>CIO&J+ykB8_Gs3%FW`{txy|r z4GZ^OY!})`rv%Nqh7~#ya~=GM;saE=kX)N zhUj%jzd@vbNO;`P!UAJuvvMlz41%fY66g#w9>Xd8M47AT6O8LFU1*%FXi zlwJ1zM+`jj^G_?P98s!ukyqqSN%mENo^Yuv1k~oCT7I$w)O%ZD3xHWs*nEfJh(pf4 zyNv#@W@WyJO~5SJ&P9j7ZjbUZDCQ<&=iBsx!)1@ujm>-IwBM;0$#_}AG!T=^qdBRM zRY26NHNwQxrhvmWBo>X6Luj7vx<77InP`q`=qyt9;5UfXs=-ZnV%dCPlW6fvm&Qg= zn4IZt!WgZ}!kj0+SnUv4!8DbCVGkWG7lt0oafO%YkPjy{Z)CzD{4kNSI9rBooG!EL zTSH!t^CglMZ@#U48lDI2&leMi_}~q9KKS#}kikt^?_ubiO$uIJSSA2u_d$kK$J6!@ zq+Oe2138bW=eP?xY{RRCz}T4N--Qy69p1{%%7iIxt)#X%<)2FxQlVYevarkd^ma2Tu4N1}i-EYu<98Z0*cf{rqL zk!4R5!Cyo(QJA&NNo<24_<0}KTAA?cjofJ$L`zZ|umx`H&3n`r`(MEoltK03njl|? z7ls)FF@KMUh&CNI*D&^F?ddOBB2S9`Lxa%#u8>XJGD+fij=$*9dq4&asU~^1AP4=q#*JbKL`DR^yw5qFR+FdjYp3}JV~SguqGrHGG^bG|hZHdQGi=tO8!QOIMrJws#*5uwZj zck}kZzD2c%U;`!ZMvajEUr!-<(%17c55d}Jo#&o4n+^`pqE#kc;914f*6A{aX>kgn zxs#X2*lBJsWCEMKc<>vv`2c5v;Na8nRIisrd03zg2ADl`<`$T)+mRS#+mk4pcr3Zf zPlpzrqa(`)N3N2Ty6yw@ja?ht*zV+?9b0gVkq$as9(!MD~-fy{N=JB4SvNbQDXi7MwDb^x33GByj%fws+vlL@ytY}?zt>!no8hmoOA@-@P zh^Ok0wY$pN zqrnQO105`1>Zq*gl##i<8QniaU*r{AIEcxO**E^L^M zd~tZf5Y0!B7_xmjerV~Mx;N6R%sW8qI6!TC(1Ye(ChATSeS~a^yxM}6IG|dofwMFp zz@1=diszTO&o+iEdS;Pi0s6Z%xrq)E5;F2_vY|a8p@DnH-AOGElG%a(1C13xcuuGatue6fD`{OfLLR4XL>?Imsw=qv>8lq~~_T)(+AVFr! z{cM)wjRT>ZN@QT77f1#PkB<cc zsuEs-vmbojqyxeg?hxHw&$MK^uqY~#m|48(w|?;Rx{^pd8S000Sqp`oNL{-6O`4)U-4tJ?d=QMKs14|Vgxp9Nil-R6G{C@OR>0* zuAK8yAXNFP@5gakc#aA#L>Oyr%kj3Pgv^+{vS(xHE5CG+jTw4I*wrS}!T13lfK6wb zypc_MUO%+u$Y<%Vydy2sS0V^QqwVl0$6`@_1r^H$$Hd<7jabobE_TR4P>S>gkkhnW z2>M)wGs<+4ks<-Mz^{)(PPfN%a2t@IEjK!UwM$Hc|9G|*IzM%QAOiCN)$)RJ`AMa? zxRL)TzJM3h8Sz#yk@L?ZfR_v=k8u`R;j#$?n59l&hbOXE*i|~y!e_EvS9`h})2E5z z)j^ileXY3Riyg*zHx%yfiqi&>;{wwGAs6Otl!vH62H>Dz76$l9@t=cyeL# z@2IB+#M2O68mbUz>5+eV&uU{7v{;bne$7jBz*7vJ2a5jTWZwZdJ1wW0$ll%#2SQyM zSzReUtsN4;iLQG z>bPAs(3|1h-2A1s*fCWjUk#1q0e}D@K4gko$|PQ8h9!gjQcI)0b5>pK>~lCl^o_`& zzQi%fc91Z52Of$X)IEDQWKh?Lw&Umv>f~>wzU4$VSUkMQ-zw%)q1XZ0{Vp_Hl({Tz zOOIW)s)lK2_LLL=zbBgPuT}`~k1kbIy(QR8)Jm`{h4~C%Lo>|$R`6}qx8K)@@6;?2 z^XC~CV@DJ9>Y85SgovO=V~x3%?X5h|DNjb=6J6c#cn2h3`QN?c?hC@ zK6QOR-Ba3qauy&6JuPLH?BZ`{LV8NjYb;+Gq9?SrH8RVmpV77|!9x33(_43HabBG{ zzhR-O{m1iUbbreV%KnrUMC}_=ST{7Ls%cePAEW)pan;i?MA|w=V}$=Nk)my_YH62e`P1n5nchr&>&MoI(5=`mWz&bSSPqZv zZM=t1)+t*maV3!>ux*bRgjW7ANm&+5Z4# z&ytF&@M`DUROJ2iLTDAcanJ zmd#Ocz-TBtg__dNSjbiQXX__PJF;DCXPDE;fIXF+% z=e*Gx(u;4oA5#Z?0r^^Ov1fTQ%Pr1j*P%_I5n&Y*HgwYBZ_QccHiTE|{jF44=&fAH zsT3*3VlA)9SCo5j<)Z{lsW!8ZL5knF1P4NS-Ib%=JmUm(xOz(wAc(iG;t<09<7ASv zpcRCIjCn`)q}34cIIT}#gVL3uf2tiou@m%Sh(x(267v^%8qRV#D2;AWe86+!aDJ{e z#|`OR(U-T2Wdr00?zSNwhDh;0WK}MFy8?e_@;Ee0(=vRVUO|$1!x>1&N)HO;hopg5 zJ&)d@BYp{fFLUD_N?cCKm+uIBNIxMRCa4oIVc;cLjyxNUrGALj8)Wc9giUJ6peDM^ zE*=1BXI|g%he3rzZlt(fVWln>2=-)Lxd9fQibZd5S%wBHKb&?8y8KaZut&i$a4cvw|N{g9C`n3AG<9?kg zc7HHqe{u^y-j1!iYkKoN_w1?Z7((L{!K8myvwV-i_ zT)$&&wcIDtiDxMM)%Stl(Nb7M<7=Yr&3Aqh6cF_p;HX`X7my?iM-}JrHnxfVO}-Ng zS+4WW{LLf7iU9R{D19*uP8d!RA9>Y^Bu+ySItpBgRzy#&$)z4Q@Uf9nDq31a^Y05+ zNXj7SAE~c`Yr*QT-<0~Bq_5CIb8~R-mCL{rydB5Vw9qM}S^a&R!9oN7uYPcjB%kq| znu1rKwbmk|TN{TjX*jxMI_M_A26=Rykt%RSl9;NRkRE*nNM4MRxys=O+WNEE&671l zl?M7_j0OF|7i7a(!f3x2ku?`8Bj#Ds?;LFVDaycVBNJY|L$RpoNPRsC-=xR)HG41= z)CK9#e_55V_r;tb{ziSgN_=&%X7M!D2@0;g181|CsM3iyLW%cx)8JXm&Fo4qS@YYg z9L3UN-+wu6V1GfYej{@%22c@*`niT!1CR(z#@9SjW>ulT*{2!0PS_BK5tx11-Ke0J zO#NQo%V-I3YuX4C_$E$o4%x7>HwvSK?}`L%xkEk8yr3iZ=nBvKj2PyV<_r%#%Wx!Ao1Oyd)pZq_d6eE~yM@%@+kbM{$%VZ+Cv%sH4`3P|77R*gl| zeb4WES%S+ez5iSrT`T9CKMRDQ-ysoul_S~ev7Up{klqK(*OeU&o%E^cnaJi9hG|fz z`E?Xr`3>EfB1rNil@}4ZcgYOdNI5$EghrO*8FQA4yJt|JEBl^aVaUmy1;E ztlE8K7Le4fZ1Jcx5LALyQNBS#JxDw7sD(RnszZ?yD4X+25tM8Y_WtRKN&O-44ZJ!%!@-+ zi_9&%sv>emM>cjlnhT&UvFIiX;oS4){}s9I$6=T-%J>?>4>QX=`@fP|c!s2z*i1sJ zn-Rr|antobDnI~xfHQ%hVMxDCy0tPFT}Nsnc=tql5v^*DuEZ(^f27PqNKwf}WCCXw zff*Mfc?e2FpR*K1O0vg({3{P&!xNtv+IIt|{{rct^V}JR9B*Ftr#6{paRfn^@_PGt z6wD47-9mQui+favHoI$Z*07Jre$|CR2u86cf)n&RJ3n$mNtA!r=*IS)pY>Buv@e%pAnlYT&vnFP)Tw4vZaP*G^H5i6H6Em0LqwjBhK zU|SC~DqMrnh3G7WioxVJW;fP#`%&SuoB^*mFodR5tH6oOV;o2;qgRxAtFPo#6AT>S zIIztIPbBp}r7T@e!%BSi!Ggc@kCb1Rls^P|y~RIJuEx<6IsB5*N#U6M06W z$)bsa$L=wc7gwt|*|IdY7y0hrh>dujrL2b0=tF!!%0iJN06M1f8I;l~IP~jbLAr79 zwRyiV68S{{G{ny=hJGv1%v|0Vb(f201D4;C<}m_5=GY&*)hSJ3vkR7KTA$SD!UbUm zCOYS)?9;j23n0M35XJ3ZcR70Fj|54kIt7|i^yUi|3gTm{{lf-u_$Bx}Qi&mu?UktXlEy?m>^Ht+H4| zTP;Jma-(`-c0er!0wd=UIlq>3$bCMzIt-;0e(m~jix4CI)$n`-#Smp24)5c{;x0Q1kEQetS-~T$LKEPlwXon>fe?=0)DS^3AD|zQ zKTA}$C@XybYqjC@bQm5J9?~B4LxYjPG~$hz@&x0A*WR}^dI+=yhk3=)DMUl&s9RV9 zp~PdK#7l6<1^WkAAr9&9uVeHYjU9u{|AukB!ECq27s0jAyZh@8proIAGV#5}4|MPI zd{?t$O+6XDlF%b6L7%!g%3e({KT zJTU#(^7ViLL-axBj93CA&Tn9k7DwW}VNmU4?0mE4;z6Y?`dCSp^p!LViw%IHqL?{mixcD-O_2d4IUuS zvM5ki_VeJm%L^N9|CJ`F1*H+Iaa9%^4d z`%_aE^m)N2o}Hn)7E>L50Hvt%U$9MUY_Ug++6MeOY5DBQ1NgMB-c@r=X{PM)B`Zs{ z8ajCo`PWt^mi>Txw5A)hA8)9clWN%CgbUCv(KehqopzZdI*?R&MW3QI-sRO>My2Pk za~wRl=E;Di$c!ripXGS^1h}21`G3WC@p%2d8!`s-5Ttuk(I#NiQ&w5bgl@NQN%)E= zP%GRm3NiR`Q)0RwMlR0&2DWngc&(q)zUsa9wB;VUK;v*OU&+RoIw%FVU5_2EEi5A} zYOmnsF>AZj{4xn_6&upEX&5w|C6!~PByoWv!%xuh&bSjfbQ~rLx(LPW9Cl3OrfdLpZ7U$$Hh; zKOqr)(60eN!-iLG;1Z8MzuZLrRZ$7SI;%`rX@0`$lxfVndflPUeru;b%fxTV@n72d zYP8NOPLYZnm%m+>bQBn_Y2n4x(JP#{_xm?NdWY+3oE3X^U znRqrvZhNEmL}3E(Fib2bWEW*#HexXQgIQaZL6iXssSW1gmvf_Y5>UNegUmuwW{U5O z^WJHOXhqQ)e>9M>bkRDXg^biozA;uLF!_%!B{_kE0`uJ{6GN(hk?y2J-$KT!`7ES+ z3+Az*`s5?bDXDs^A8U=+oHV6cKAS6A(K-B=ZJ3$MrMr#b!8GfVs?$rod$kai=}C{J z-f&cz^Q25s5K&UP73Vh3g2wL=0B_5YRIb8)c9wh4fhCQskuKWKD#@!f-n(pkL!j(N znIcO?cNu<(D^~NFIPtAeKx6VsbMb8M?VA9qLHD&G*MUF78`Zf(8x$#^F3IrLxquAj zDr@G#@*1-8O~Ji&K@MBor91$EmTr$nx7Z1MtB3jpE(Hq@MhFdBS{UWm6A6wT;bDNi zs~j(dhH4R;xsZf%(P1xF`aij5bg8{&BvmGe@=!;SOuN>ci+x^NM<67vWY{ z6=~2IhC>o8)W!-D>^w~YDugndTe|1+Fhmafqwb^6`JLHKw3aeM%jx`i6bU~6BbKY( zMxEf&{W(n2f2xl|^ z(MyZolt<=k>FQauif%%nZzLTp*5qS}s2MJ7x;gz$7a3NxBT|QRJ++vb7X6MP74IED zWc`Iow8>w*cMs-fboJujJ-Z}kBhGO`6cW1z_(YCXlxh~FUqn%S)LV#|gUUjc{?AKZv=gag;1@-l_<>y{@^2FWr3Dj;!uNpA2aapJ2qiV_Q^#5+NB7!f-0>epp@))c$X;^4G^-Nw$w(?{R@w@~d zc-DdO)7v8$^c^mNKJ1V!+zVHC4;n72{78$77MLC2GV%Q1Gu%C(-qfbN?}hdiNK%|6 zJgk{qT1@_6p$U3|2lnijs%Fhjf-qkAB6ZA{$*Jd;bm<`ePbO;a!iC0VMhhFAH_QI6Nx=DI14H;d zgjBrJNe;LHh51|A`VeI;a~c~cdQP{sSt6FUHrZQ~HcFqFP2ibGC)|GMJ^a?W{sLr$qQA>ltF8!R8dwg`izdb~R5c1C zqW!Xgr#^Um%xg$T-zN9JL$}DtP^i#S`!__miYWW-@h-<7ETf?J0o`_guTk*^Qu``Jx{iAPh)Riz5L=A4gnWHV}_fTF`pd#>!Z#PThm zoKLkbuuM=V=dp)R4soP2OfdhQf=9T zJpMf{?hyHl2}oMZ#1G4U>ns~@_SILyI`_oj%NDDAb=v=_r(`qaWk>MRHcHaPMH+!= z4}yB-|I$xATv)JDe(_E^-jb%y0>umI8|<;hX+752GQigN5tHpt8*VEP)kBY-!|#{w zkrM-yvca7d542x!U%rTi%M&k1%%tw0X(d>8B9EPV!|}$-l0myeue~ z?!#}{hoF$IVZO{;d5jSfv*Rn<7XQ6Q;p6MchI1+20%}72By`eNJzkz1D>);*O6A)r znKSi7oR}5=bFLf7a{!IJXnaM&5Ki04AL!R*M%pui_$c((Kd388;NrGIYWK}IVYSgz2~(`rPv^nnt_9q6Q{H0fuNqXY=Lz&ADYZ>D z*Fb`(tCY;JHf|EqN{A5uLGub)JasFLyE`>6SuIL{4TsVwLYEZ=>Q+UVfiN=MuaugN z=aLP_j4qb&U~T?-U8%z|wj~<&@pC`hxd=ZF-dmD+WmgcPmFE+-mogh|C{>OrlZxli zOln&X^zQh#`W^_BSRq!v+e8}>UL1qg=4fQL8ZW*asA05ER^oGSY^pY9jh4Q|PNLoz zGqle%%B-DjOb3iy=Z_G7#@}?yEAx87^5P=`GQj1bnNSGT=w)Q+(8NQ!y~6Ica0ABRTqdhS>+dklHuas35zFfS-k5oBj9Vz-rYIuFw=tdw#x zOGn0!S^p@ZcR!HMaDkOB;=P3$eD?u(^tmeaFn!MVRLv3=!qyp(t-3#{M%V)as_ac$ zL33ywR)%xpG>AV^Gue+7U^45=h&rD{#*Rk@ovjx+(-cNbg%c2DHJfN$U6)X%4}lDH zy6Ri6wCBPf@da;NOXq)ejGd-#8EVT{Kp0x%l`&JCc41va_IWC7V&CIzG!W_fL&i@o znq6J{EbT579sy~oc!oc4VTPU!Fs8UzWDGY^=BXA2vhwhvx!#4hyOV@!EPiUr((|RnH7_xwU$GU{53=3=ha)Ion{X1$?;C^f64#)S9{& z99eqw5wKj{V*x7Mq6&s0OKR`|s3&=t$3{LnzLs`W%}vtJ2uvPX^57RxMShiM)cd}4 zqOUH0#`0utZo0HMu05|#S3`%B1>%LAE*zysENOJU2^PHb|7HJM$n-`}-R(+rEN*_A zSU-Ees)7z6O;T3~o@8_W&58C3Aqc?)JHE)?0yk71ajRp@t9Z=|dJB(RJKTv?05cAq z;_!W^vhdsA4h{lyl?x&EU1M_8g(p4g{!?69-Eu|1r_ zQV+&Mq!Mez0Gd{StP0l0n~zM>?ka4(B>G-DbiV`7Hp{pmqXwf~?&8M%XRsh~_NeEN z0)cBDeVG7(x2~)SdXsAI;ED>20q@2n0xW7%C=fiajZo-ojqs5<@@X_)dX5Ga=2b03RGJHb!ukL#u%Q;zu54XV1I@i z4LMh!jIc@j3Q;>`w}=uyC*);d(Y%Rp>hOyp!({Fa=6zHi9EzZycss=4=Xe&NcFLe7 z+sKXINPN@RuikbXf3PD{(yFpU7gJnmLn>Qup03_TL)&4pZl>Q9fyLL95{HlPBHKf7 z9%+_zPi?7NciVbs+@rz5Eyznob~$)_Qdl+6U2r=|r)olRZUxMN?vV|?jpGW7s4EQ$ z_nXlPZfe(OXnd014sB0`_*0XrVr4#TxJ~#cQLdURHH=MX$?iSyi?9BzrHNGrl6IEs zo&%)pYs)WeSJscFtWZO5AyC=D_&16jQ6vO58emD&+7-ed8?crKpi`Y|hh%?%zJ~E6 zEnR}UbrBd!ccN2N7H5dwcnf5sl7G7ax8K5!kf~$#uP9Kx7@L?|*L{lxI5i-ov}CQI zH|%M>laNr22_ncSPfF!5(_HK_w;F>osg$-%XHQizt}PNBIb5wzQ}lp-08c=$zr!V9 zh&yT&lwI1!Z4OlV&5c6y;eUWy?-jKkc<6A?x*1OVSPJtvCJgS)Ld&scFQPWj11W)v zj3@mZA7kn3;`k@x+Qq8;uLH=e#gQjQ<63H)8oQ?_mN!{I&}&LGqBH>&c zt&n{`Y)>zq=P4SnHpX}+J3q-H-zq*M7qlCX{UCtajjPcX1i+IT+_Jr#``Vr5R6S%( z$i`45H9aG=W~STLz4KJG7K9tYuTfHt+;$vaq9%lJ%0H0id%))@;<1GDx4{C^&#}9b zGLsZv`}p3n7hp@B*~op9czBZZ_|u(PU&*%qmr{V8NBnL;)$vs%_yWRLO$O%E$G5Lv zC^-bV9WFe=lVjo-16 z0WpdCN)V=l8zP503Duw_M)2p`p@Fe=Iy1;!Sno} zj_xRgnzL$LC>kiZsz=%T>-WO<0|P^DVxDO1STrEm`X>^^?S<1M#T{0os|woxt(9@s zyi{RC{}`ICr1_=&P2w6I-#(etfOQ&$D+>_g3)joQ0`obwC>W!!!rT?D*wj?)Ld&$v z!VcbA9Dv_BuPf)a;uCP?xkLy1d4}__=W0u_PUDlWFgum%2Vhm-tbFk?sdWXJv1~aB zLy_zA6gci~3NY`Z{vK!X(XnCC!ffKJo7!$LGt>m$-`~lr*X=-#GBH6?OQ)dE&>Z4z~54Yl`$9MC8p@ zgyGv?_{nhJ|4lY0ocU1>q#OYir~a(jt8opw`Oqyq%hd`A5!i5qN1czb}>cp zE?^4@wRUN&0()66ZWjo|M)xOMdu}j*Vz!o33y6GwO``I>IdG9`Gz$eV6?XYLu>Z=M zmIv75pLV$Rr7@*JLN2B-zhbvyIYV^>!2sXxXxF&|YjyqPGtLXoZ`W+dejFE0!R!QG z(yb?Zm@*~TmI*gxHtT<|uZn?dk>kY$HSmDmRCtoQw3~u2p}>21NHr=-Hg3_x#DNYG z(xk}U+!>La9@I4IP@BW-xD9dRkr9q;2=XA2g8=2V}(MixjFL+vO zGIK;5E#+pcs11+ADM2$WDYJ5xJ|r<4bxD@uP@St*&zqC?fOuWz$Vjq+<*9&x)lA-Z z9~3Bf*io4nB()V}0UeFkz!d?Hm9$*dkQJ#IY|}uam+ja5+hgihuVc3w^28&+62a>c z*&y6~9a>o6#YzmUoU`|Y-URt~bDkj(eoQk1LV7|Qfi*F&6VI8{r;(m%$7YMX$Bd47 zT=^wqTbF=86(f5=C={LGWws=ik>M+2d63`Lzc)lFB=z|aKj5CvJ3&J>klzoIY$Kd5 zdtK7{Vrh5cqXT={jkabE*kg=NUx`u7Mj}AHKX-WD_Yr9EU%# zwJl-xL}{7o0&fsC8V_CW=bq!5po)@xxewG?p<8h#(C(RIPRLDu)(K4dJt51oOXiWV z(3cdT?>V`hONEb3EjAUQAwgeSfjsqeeu>i99z~6ZX2~I^xO!8JXdq!wS^|8T6TufO zlWudZZ)YdY!>Cq25EUl1NXJv1x}ZTCj^36MGAot7&(3)$5cHXXXd#r)wa0S{`mz2i zJkwd5%op|SfY6`g%GCLg4Om_cq@9B;ntE%C-F9i706= z$Zhrbb9H@A=NR*tqBeyMmJ8LEF~DfB$C7CXOl)cpbvZBD}`bAZtbIDN-BO? zE{4)ONFkFyNz+6TkdZd6Hl2T@=@uu17#5U}8yvZMCzZ<`4I(1>>TWFd(f~uQzfnT} zQ`$=G8S7ag9Od)w7b&welGDaeHAdP5MK+asH(aanL9q_(Dr@JePYPaahFfy$4BEU9 zpQh$3&PMMuf!SmLmY*vqaoW99F+4=w62xz7tJC+X;Z{4PMl97J;G5KD{263im48EU zDU;Qj@d_s5dU~Xl6w>BSxUFFVs`<kHg8ls`y07|OxojvvQ1adqqw;;afz69XNDL}x-E`6BVgcM(>}qia->p9@RiMiu+y`Rn_1U3@dYG(I%D+p%Lt^l8G zgC~5(bSg+SW@<0Aj{=5##U5K|&#EPFNK&y>V4VYcKA(uwB4K4=I%1M3yOq^R<@voJ z@aSw+Y)In3%xNMr4iXg06moVZU#x1RwQXaBF^*$xLil-9({GqKbRyJMsma)?m7L;t ziV?IRLuu5D?&zjonTu9lNOke}905#a`s9fkor)F5Ckr@%tCUUtrTAa*c3VQU)YV$co%8Qa(u6ByS8+8)!t1^ot5;^QJt3_&TKr|#j{ zSy@G$xooBU{dIdc+JUMyoL~%M^Mg0gZ9)&WgU9=O_2H*g(Fg;ZURx~lR`_@ZV!%G} zfB%>OKD240G|QDM)NY~;F>31hG(sRq3kIVkRje?0TPUkQO4}(vGa;}` zMz3S)-2|Qr1~LB* z42W^y`u-7(oOQM}U`1R_+uDUBXAp_$XM3(}dVQNkO$af#a^sD@qv#u)P-)K@JwG1) z=_wm5vPGJsqzJat@If7_(=D|TZ~P~rl`dqR)&5+{YJNdfCO1t9WNB_ogCw841WZeHyAGrX|as;L25$L>3wxIiRe-)mJVvG|ORC?q!Io8838!C~T(NDG+h*u;ZA3 z7S5{~k3sH3!-xLe{xohEl}QVQ4p?LS6i=>p^Gk@0x3NPg?HPE%f%1a>KN*+MX(OSZ##@+`{g7)(7NFB5SqEzkLR#&_yGbTsRZ< z)?BhYZ%-oangPNr9=oMSOpA`nWeRME|6CPX@(Nj?&?cf4?9c9LC&OLwxKG#z=VZAu z>6?~qhT`Uu%$*SdIL3Gb&oo}yUJ*hHuWSc>j3mdXBKA=I&=igOQbODcO$1d7q%`Nb zjBB*$?StV3PJ=-c8y79}ETx$`&9;a_gNseVCY)T6FOUTIT^jiW z_Ptv3d^)t)s;J@qqldM`LnpOR${1JzLFQRE`rgcJ6e4(rWfM$Kfo7e!ON|N%82CIM z&+TO&p~cAtg@wX|6KuImKv6}Syj;dJ4EWW#c8TOVv^aANwsU25!a%n^+#G#Dj{ehd zI*)b^-s(Pg5c$QPK2un{dr<>D_X_%DKfxWE6%wI$O_KqVeqZe}Cns0@mPST0(}#r6 zs)%Hm#R_!BaF9h?dFZ>IzuM4U9_uAF_JMO|Z7z1oEf938JEGi6=V!hW&xym5s+SnV#O$t-OPZM}=+<_sHPDvsb1 zjh8Yu^c=1QL$`*IAjwNIxi_9!fCw>rv97c=+yxM$x=G{;CkAZc61fgFw`tdlnk10&#X$(wQTYpGMxecad=?$FMJD{Dn2ZiTJ6*yoLI{`qxKI{|QTGTxrAh zBgm7q%>eAJ+&3E*3%21A41)U_dlnM;f`GV#q74q`4r|Zml&!(R?(2yhDD-l?iy$Yz z@%SOwH9v|zo_ls*ZaKu-2FJs*gp(>rkcu>6;oXHrQFr_MUy=&>afE=9hSa#2f_c== zJA0oTJ)ZWat5d2rjMak zzVsA|z3y!M%JK+J%YOpTRu}{w^-p(-QKL0rg$7R|qgi%cnAdLrqgs~I1e|n2^e2_R zKho($4w9dBnXCuQC#0W= zNIC^c7WVR%W>O?nQ1?%I5VjN89TfECZnj&=|yArom`V@a3eqTrB^{1z|YX5I`=#?A4TmUGW67{V@-ePrTcOWC-eKpKMh?u)gr)8; zA{H7|TDGnODpLrwOS@(|wT`2xL{{azs~FTjqttME4njaIpG8;%z!KHDWG@pSCqZqH zn_jg^=0^WAYCF1SzOSnzN|DN*a^%3Zd*X09XSw2T1C_90^A1hPgay2SFdN4WZ-tuq zDlEhY&j-R#`{m^1z8MS&$+VK)vOH4eWagX^r5lf*{Yr7p)$F*IRqx|*v}I6W^QU~Y z;$KJVB7?KF7n~QD3P;tRG8F~A9;6oX@@p30lv&lwe|51va;d?|alC1lX*c>wu>YL` z+B=?3McfhAl@Uy70$5@G@Tn;#(s7=D=KJLO12*%JrJN{MO6StJtgy)zXz;aZx~qsQ zx(M{Ry>q+KICvG_w}V5o22IF^&30zn5?-1XqQ8z5wkbeK-N+2Jg?CRNOd`y(Wk*5^ zjCt%XYv9(nSqvY6UvTA%tJ2W-b4b~o<m4TO+t7SgD~y11Rsm59oOPK9aPKjyZCq~@vwj~4CIJ)B>26PQSM0n!hSKM?bY~; zoF>EGIssd%4z}9gp*ddfd!s<>paAE>Ag^Z>WjvmYAb1{r#;=N(V5Gb|e-NCeL5kCZ z=^C>L{052RsxKT`a%wEUBPI_%`HDzw#?OogHafa%xcVg6X}+L$55;%rkAc+VaUmwO zB6ZdgeJZCo(mZ>)#1kZSSs=jVSKyDjJ9oTqXsrZ(=*NMR7+#KTt9`ptoqGrMLtiI*-ys-^&l96e5WMk+n z&jw<@IA!3S=QFq1Rc{#8oX-rBGUTSLIdPM|2(Ss57U|zF~IwAVmB*2RshiyRc7IC9QAg}Cn+j7;1d#Ds< z@gFVTdG+)ELl+m$m6_Me(a6AR-F5=HJrL)}YB~tlzpUI@Cgj?opzayHOZ2!%yyTuXbf99`LVHH%u3mxPDTx!?z{ipKGw zHGuqjscSwYk$xwsr#T(^h#+ z;kB1Z?Hda%?+K9fSF~$eh_^@_YN2Uaq3#g;w4xB^K`Z@90K1~^Ba88>am;Gx+cOr) zBd*F67nK{gOOOzWyHpVwZ#oI)6umU?El$#-g)azWG&9pY6~drn(t|oHu&DMQ05lbH zonmc~7-`WESJjn>BKf1AuxIm>xHh<-nfKA1k~y4xFe|D;SB3XM8YDX@>FQPjtpjtM zIzF=8t3!*){73m8CS|?1wk}zmLvvD7j~<)UR>R@TG9TFfzG%+(bpc(e#Htxu{;O2x zcngbnH`EDyV&rJxkHMxH4F5`J$L0oNT+e;J^3&Lq-E^4umFNRXSgxY-4%?r@fXV z1rc;*q0f*m*>+!k$u-s?c3A`h&Qr;FoAqyzHa$-L%M;0f&BfYpEh+cvP%&k2`3&kT?5{7= z0VaMAU)_wnq!iwG5E~;?2je9;SxM{i*588r5X67^y5DOoDp6F_TJ|~2X${|n<5XXJ z$?PIH3n0^+q>5?I`1qE0GxCfaszj0T8rS=P>xqNQPMrO^W@m)2`L{|tYgwO&EEboW zwrIpvI{99 zSW~mcY!HU#f*5hp#o%)LvF1<;1@Yom)5^f6Hj)8!Q6a4LeLNz>`?7sQ;y#LqV9j+; z51aPiR{POPZl6@rUMvJZOu^T(^-HduDZ?lCO_4`0@cMj-<7P}paX>)r*zF`o@0zeT z^hFr1K3L`u!?|s!)&*HR7t_{K#}=(wbmKwciG3QO1Gjlyj{hUR{}ZvNg3@~W$FN(1 z#rCm7R#4+vGuDwE|Hc@%bgsgbfXaJrxs6l01XZn9)d6uwU&wH=>)5mEy0Fm>I`3MB zYF(HD%;fM!3YI7r?86t}1~RcZaWBN^mJcAMVM5mx?oq`aT8WNckXD6BQ+^y)B`S`{ zsUdibDFY#Zg$!nj*Bg>c_s9AN+<1JDdSE_C->*?w>L$wK+UU2D0T1Absf@~!Z=9pC zWaR&kPeF3dXcmZtN@4AG#3rdg5|ENJlwp(QE&ei)r$=D%dA^)F@Z;W(wBYu^nu<8r zX2D27EV`om1B*W(ca913IwQ7PQEhPDt2nXMN&>0IVc~4bf+vp<`12$Bs6)~(M`Snj zaSW!W`gaTKeq*CPd+XwQgX@>!Xyu9bU1TiB&4EJX=jVq(Lrz);}+avL7{6$Iq=b|FZB_Nog6)Q$3?5QJVjyfGpUoo zVBQCa$+6z>miB&Oq{HW<`VUw5X#CQKI}m?kcT4VK?gngp0f%9Hk#>*BKT%4|L|1v8AZ4Zfcqqzhtj{-jE&5R`h%oMc+KTONr^O(MO96n>R( zz$9Anz5m^RY;z(RLw^~tOZs5U(mgU-i!IIWw=xe@5l?Z8nE8RHiT^w zpCGE1|HMlz_`yx&F?qJHFl-}i+yDUF)s}ARfv0$KLK0f9eNNN%R79_w@8Q8Tc~F2E zWN(Yh9689|h3(C10$n@OQ1hENj^>DTuF45jU5~SSN_7-khPYcLy@2*A)-)R}E7NO# zAh?_8Xj{=E+<0lU+rv)QbtVah?eFJd$P2F|eyhMSP?*-ra@*eHhJR`aubWHPA83|I zPK49Q(x=)2Bq>o~KMZLhV7`)vcSTA$qr~^~&7&rpK6TjaB zS%WfposL2KZbpK(lOMmy0Yba@NH3?n0Im1VYlaWkQGFe7q|&3OuN|6o*z^fs&9`qI zaGYx0IbVSaH7U&`(h|@=+`gW@O}USm1zO+`fLM_aO(AGzC0uwGkQWA48wtspikQKX zfW*!|C#KKeQ;BkJe2^qAYU_sTG%BND&PuE-IcMJp+7q*(a#9AnI)eX? zY`z;Xcv?K+DX*>p&123cR3;jh9V;?aQJJ<#-60w}t4KA4#+kT<%GtHJ*)6B;X6mC7 z4Hv5ZDp=7Iw2GDWjm@A4P@ax|UbfVJdJe#4Hcs!O-k z`XjEUdM*XWT;tF~WRZnq#cgH&TI@BFv~Lb_ZlBH4XEb@7s?ZjQ*t?Tem5O>)HFsbd zIQj%fhH(0BonjtnVO4$}3nqWr&Ofe!m0U}|zwuURY$JNhz;mkP2eJ&Sj5MMkrG8S} zUI=U&%yJUbhn&AQ>sMe7Jb|Z=NmUzB24<6F4-zQiJh6 z8%FpGFkF2&e{0C>yc-|B2DE1q@xa2b^E44^F@2;W|ksKQ7^$0azmSeqFw8mzg<9^5i*g z`bWw{uX>_VlGK>i*7}TtvUA}cPAHX^s=!5K?mlD4rCy0nhW4~*=cn){+L;~oQ7%WK z_rF?CJhrF|yrS{ISd9S}G8-`ot1qEoGrb8{1#XJa^z6oz)nT8*J%z<-EZ1tM2={|$ za2G%{>+$ikZ0BbVb|aINZwUVs0{Zy2P8p~60AZu#Abx>`jp``Pp?wwd(Jb=*DvBjA zp05Gv`>7{SU@>@m4Y&)v1Mc9wYq;_=R-t3XSs#^ELG$cUO>k9st2p`g4^U5dQ9n*C z!4ZoG+TT=0?k{h>KowQ#LSy`NHV->WZOf9&r6p&riX+D#9>u4LhEey7+Qxh*FEo`M z@V>PV9zkt(joZ#d0C?8_F@hQh=*OIFP#=5#g_zBX(+7O;DLdCkAjz;D#Q>QJoUFhlB6*lRHoVN~Fy2VQx3G%!BUdi=-M zWWCp0yWP3^yk-30dmlCIWxBy)9vENT5WU0+c52vpvmCm-qx769z`;y;m@n*WpvX;T zx<7#aF5@3&R@Kh^a+IPFJnCwfxFf`${?+{%O1_#$AL3n*U*Yd$Eb03@X>HVo)0oxZ4fNu7-VU-iO>BI`o@( zr|?e=?5cr(7?M;Yo>^9JyZ$2M1+HtfHIe6krrcAW?yFgnmMF|wYvKO^<$wB2F;?zT z+UX0hI6Mudv}t&yUB}h`{}akyS#9-{FawFE}EX-`J5; zXho;lba3)>FwG~?JybaM7?Ry;;T&%opA7^XsZjYw9FX?szG#&li#@*zZUYD6$eRg9 ziG7Cyh7?BKR{NO(27e%|WaExc_Ec57n1^cq!fH+>t)MX_k8l$sRj@RPim;k62GBO0 zBnovI~0b+V&QxPl~tYJ0r8~il$I;0pj!U3&hygwqn%<{Yr@0Z zpi}a-T+91meYwoWWXHqLY!Akj5}(EOQidXsEjS}(P7H^zw*TB+CAgdDi^Mn>wG@TK z`w&)_#j+sigKc+_ki%}?pyCuCK|@TG<~Ja~yp)0bx2BYZ@2IwAiEIzN`*IO3jMW#V z+==b-hORI*p#Ps-=^Fw8|6cKTy18p)Ubmj=q}d2HOyiqt+%)kImA&)z2cF&pkgxU_ zV7ctikO1&kmq5l@u-=T7PuMB3L$ZK(<6)<0BHKb~+aZ(c$tKih?_XNt?SDVjBh|}T z^2c=4eRgGr=9NnJBoQD@d47!GsYhrIF2pkTqPw=T3V@My!mHM@bJ8XKQe}#3>O~KB z!DHx@vPEY)K5Sbl1MNDNhFf^bPMEZ_ z0go|fy^Ia(;^GRno<$YKSnsM;s zn3GYng3xvB@UP9}(aB4x-U1)zwf<;Cn!a~@i6G`dq@g4yP!-Jtes@)%^As^|s$zn~ zpkcWx3WE6Rf7h$(Dq?k19>^-OO>hwE3zBn0thN;FB@?(<{)qq1cSZM#6MCqS~go^cYG7GPp z#MP>tU<2~|qy<5Be{7tholq3@+c)>RlMgat>UzNC42iEhf4B-?128l# zy70QtTXi`he@xua2RoSp_sgXj>YAiBo)(FU^I#jRV#|Y>{y3v5?y@SNre2j|Jkli} zN$h$X0EK07Pwqlv_WcdwrkF&TGxna!+oDno-#h`;vUh-&pXTOzPX|JqU*GSU_{+I; z21mh*jcm9BlyV7gKn|Cq9XLghKnl34h~gkh=^1{w@PonUm@xnDty!To+n#Q3LUu)# zaDZMBz3R=fM7Js=wtoc>29_W{^BnDm{Uo-T*l3?;mJukeKOuu<>XrpONYvcAT$zEV zxR0mX!BxU_!Fi)s(lYq1W5cb%=-D!@KYMiuUDZ~X z$C@0eNygs}ufQ+GAN-IJ(0&{H2@?Iia&i2zKwz^^Ux|(=`f=*f#&g-jG3Oo9Bw=d# zjC(oSUaa-mPK0rP^p5DC2YxF8*U=3hBn(8P73Q~r{WsTRpuwJhJGw5hWd`WcifP$h zpIy3IP91PhMh>BJ0bq`%-h#U_=LaliC!~L7pmB?5xVh`&VnfL)){sZ|G5~=Y65SW% zZ-3lZ{(>o^z^Ue$t*et7pJu3g^0#28?-`f_%3!b0hz7AEk(4fj@(+cR_?A=3-VvysMQLa+{On{kBvg>goQP8_+^D*MbQS6i|X7F9h>! zq&GkH&>a%3yTFVI*tZYpDQs>dAA7gAfPwgI#!Ogk>AZ;*h}z{6Z3ehr|2J#v9`%)l zk8aILeU-VFcp?xL+|-%In~ez(Oj@IuarTy4@ijT^Wr5&Kj(rrV8*=UCK&a+#)o ztXsPgB<^KAt~87Q>3|sxn29KZ!;T$8_R-=vr>H)^S8YizcjL(|CHIdg5OHDUYB-Yn zv_kk*wt)|~B&KQe@bsD|T+9kU@hF{v3w`geUi<^{UW_5DccAm=e=Oqz>+H$2Q>eB! zd>868^m6LjBecis1Pia`V`&~M$T<_Y%U)4X=+hSZ$kRq={=XVoQ}W+rlvbt^g6Mf= z)eOCNekgGFIHRO@Pc3lm>Ln1-L;H!|WSaHDj-tYg-tbJpye>_9#nx3gDu= z^AFhOxm=l4ubwct(R{Dtm20z*K%K;@fz|tPv3*7Ly^PP08n+=Rd;tZ!5`R)|t+ZTj zlbcZRGH=VAtk%nBCVswF=?AW50qCHvaN3F1KAic*yOnaA4rBXG+Q_X63)!zsao9x} zjZ7)(ma7zg2CeZ*wsAl?Dxx4E){1$-Bs{v z1jJlDj1a7Y(4c*xlQ@vk5W$bu&$lMUfMQiMDpf1XEnc!hg;RPOL3$8LW@=2AUxk-{ z!QmWBt`6H5ol;pc@QI8hpgyf(C7PY=wKQU7%xN1(P z-_)W~LjD*bhj^Emt-I8i>O76!9^!#?;Q@mJJ}rFWGCT>OH+M_jsLnv&u91R6@lnVs9Tt$?q0mr#; zHY&19Pll@wHv%`JEyE*3VEmOn7+`WfbW($`wR`1)A5qRqijh=-lO13_*v)-=GOSb1 zyp+1J`hBe9@-F{IF9Lx4aO&K^B-|^61A!v^p1c|-iH@bgvsmX31qIpA@lItoV-odx zh-&wuT#K%uPdZPS-AW~*_Nk*z9vdEK@pT0qc_3QoIy1XSH4-4%#4k{+2xe#QkJx#V zFJC^`2?xfw$3W*_>X0{YWX*Pj9SEhma$4|C{L=Pq z9(1B$8+z6?bNIuQXasTw2tlMgovpxhTjOAw=z>#G4{nS#-PFI{7yPd1ZbymW_jOjt zT&|h51Hp+{0%WZ^W+I#h&Gm*GlYH3`Lp2hh+{or_tUtkt;$}+%V8s+EiApS7UEB>r zE|xHZ&lm2GTix@)Z=2fA{!y;8>0GP3@~#1PN&UmtIaJ7KMHu+NPfN`(Nn6P;EwM9~ zrmGrf(}C}O74RBVM~Xsw=fKV)&Y)U8OZ>KaJsmjeQXXL?%WFPuuYqN>7ITn53GsBh zyWtF7JYpFoH3w5(N`k!keGQMe?2#&+ zbAjT`KK-djcG3y{+qB42$Z)F}aRB=*23!dTyA@HR*LCwk1GJeZVPi?N6npo|^|BjZ zdQ;bWQXwsQ`gL?@^BLJTTUz~QHmCS~#N z3(T+mx&&gxTpu9%OhG0TqX}E}RT=7&J~IDJV!a>!N&kx_kw}{aGrIoI9F`EmLdQ}| zqy3>?)82ty-1y3{H*MD@y_87sSTvmHysxu!d9mX=Y8+#Yi;z`D+KSyenMKSDMf6jE z@L>VoU1F}$Rr)oTVTWHk4;{XDO?xIgM`oK8Y{Rq`l4?ORFf-2KkJ5D z`7Vb&Qw6n%Q~jhMkuk4{kV`|Jlfws8RmW%pYw_c>KK(={`!q)rW&J&loo@9^R35RD z%8rj|uyO|%1bpdq8DM!PCljq|Dl54S8DO@wk}e?@u1k zPVf0SFzS#X29b?AzCV<_reCQ+j387=CmADmmpiLVv@keR(G6Lv=F#*hakv$Anr@{^ z-4EW|!T3%{Avw*p|I)qxKs|6n1dfbszQ%K5{fYN;zJE!2>@L)UCUn)vl z?bY6j)~Xz)__o5%-JQ)3Dpj)5a;B-I(D3b`miT9ClkF zj^ULvnJg$&BBKa6&W};)uei<3Gn7B?=h&-ke!pxtBjK-Io%J>wJz>S14^dsRiCTTu zEya#vpgXNJZ3!|jdZz=j^h3p0&tWAYC7GykgI{ME14oKN-E%P?En$zz7DEXoNJz2* zumMo&WtKa8V8lNZo0cwUyxrs3rf}O)z+fCC!1FNVxp!@RGItcC6RaeyfDT#Hxo)$D z-#4!MKyIMFUdQC5Wj2o~2NJH_tH{7)sMv#Q9Z={{_JSj#r9K}RDT#zJIRER{yqk%D zx#HxzopX@XpA!8{doamkgStnl+CqyEoE#BmX!t;)^dj9XEzBWDSfj6MS?9aykEV;# zoOReo3g_&eJ@Z+^SnRrNDb>`RaM(!bp)>@$2V*FGoAxv_g36g#uk0SqP+E3s54-f>5 zo>=kL9~wwTI!E&1A$S^}(BDZXUY0E3Ba|GuyK~4)l1cqnUkW(nee1TWNOa^mHA?bF=q(zv z`aH&%z4#)0f1+U%ydQnF*V$&!MPt8%_9M1+h zJv>9NBqaG@qAWcy;!QPO`8esF?PGaih@ylz+?yB5G$HiDIS5CUtJrblwTFb)ud0HI zD$}8TUdLvY5bxP$agFctg&_?VxXU#ti2RL=9bLqjXln+wV~vH9IOkHH-&W%=?TeX> zXzS&n7dOCR#kqsCQr$kO7UfUaty6t?uO0yIQWpN!#wgCfv`WPFp_+w&50+p?^SJlk`rUI2kTR>Hx;eH8LkQGED3j8*HgP$x2-^-%b zpVC>LKM%VF==6$*P8N|?0CFC|R;-x5``oFaXXY2+-`EW*W zwT-OI`)B?cPlS3kS?9owR?y3i?>Ni_u!Q#$@Dw98qC!*SGjQ`$&) z^6Ir~MtQaz;;TjQj6NLg0FDC?Na}{PBx_A-5n_c+&3s3q+w*QB@fo_6NDvm5@W%i_ zNYh8lVh?`Ufg^Bs9>-K#Nh^-*4Gnd#|VK_LMF|2kCmH;znPCnv4+ zDb0%8Lg&5_H-%u?IaJW$n0{9!aqHp;jH+!?nv1*q#4j#lbvRO6`@N)*kt!ui@B}*jQq7WO^(6y%o|bdMQ^3^1F@z|OkgJM z2`4tOOp#YL5XX~Z{Fme28f^y11=Avy;>u>JgRkZ0Gx?YNrzON}$d)j;s|izr3`pxQ zC>Gjmj_x{w%PZ!4R1bX>8i@%aoEkfpGH}v~)T`2RaajKiyJiCE;D1$|I_L4w>)*CT zz<`0a?XB3a=a-Q%--NQ+M}_<;PF_h3$a$aLULJ$p5u|<3dVt&}q=12BhBYnDX4O<_ z>6H+uf~o9JRO^XI@r+=Iu2bYm%K#uv%k06Be^KPUR?;gCZvBv`5zZNbm}LdCt};tZ z7I9IXr}YIhg!M3}e@Gg32=E!@*Dmqz1O;^HDt6^VF#sQWs8d;*qeqq(G)3P^D&BJK zA|1v-w#{5zPG&mH@>y~MVZUiKX{59HBJv>ARj-pYWKz0>CnYt>fDtMEAAb}U+21Sr zB|W+hWdgPq?P)+9zuNd+$9}>Od~6+tk-I=9YKxShhpvk}{P7;86k9%%DQ{-L!tf*3 zOm!-;nVH7}bh+RmoZqdEdc9xC2YaA&siTd|?7S9L_=!pVnI-hpy=CJ1Kur$&(Qr6d zQoCm>KEe}dy}=Rtz@pfu=b5X3GVyzfPXAqZQ?AqD4boT^A-L&-z1F>{4py+{#_a&0 zg9dI4iqd0;_t8bZeb#N0WCnq3@Aw3sG>yW{$u%tjU=DG=-x_r35TS`Y2}XunI0$JOswqZay0unkZ+FV%!MtwZ{KQIFv+Z3;7+sUZVD09 zihA=@sbhYWuf@1Ztjs+XhUoRNurkIkMje zKjMAJs0KOAYM-$Gw#|RxoNQJTbj9%zGd+UqwWzqWq=#Nhl{+5eTGizq0Cs*x8kb&* zMIY@>2!b}r*I>?mIO*)*vRE7Fi66?W*h&KY1eYwkS>O0)c+|3$Dc9EQn55Ty;2rH- zD-}5(?sR?E)F%$1o_eRFz;g7wUxEw@iX_lx9IsfWRGWZg%A9syS)rGXwJQRTV(Ng$ z3!G=S8#&_FC| zC%gKTs%T??LU_Mf*;NFZ!e9;E`BLkO>u${uc6&W>vCm7;U=@P>s zTAhHz^<#@PmQK%|!9KpXnTkLmPa|}Y?)ZFGgD!O*V^TP(#+oSP_BPBYw4?8gUFOdW zBAr)N03<`Jvb+eQv}$%UXzJrJ;m+<#K4CJLk*SE&2`IM`U#LaQzn5_r{0$9})OVw~ zJ>b`;fYTr_q?y4TEpB6Vrx8lM67NvZuvx_ZJY8#TbRtuUY+$TkR)O5VVgT0kBh>Mo zpAqoQrY>p-DMRuVhn+u85F@%n3bQ;+nN;ugjmKTUcXs`eR0;uwu6bkYVI%f$hS}{` zzQM98VZk0&fPkl;J;|$dK+uW}GhP*RV;_L~Om(X8lO#8B6YKxYtfUOk4`lMg^Gmo+ ze^Q_Xhv!8_TS~dPo`xc{#&4IOLoqRJR*zqTrNFRz{_Z%_`7K@>t^l;OL{|{H<^3B` zK&2HdpGD1%ax+vpSVqfyJ4z)P`-dIV#iFABZH6T9^<$5!Ubz@#`O~XQ(#h~Pl$ojE zm%FHpR}gK6cQiHVf^?x7NX6)N9{^#?TYVqess`MY9MCKu%u zd|vqhvmp_2M>+*Ab_P|47WPeVPBAICO-^Ps1D7?SP2#8AoN@lh5k`++Wj&-<=Qe7z zAZQ4_{G=?o(2G0{4*>F=3C;$i`)zD~xfy6vatWy&m846bkki&G+ktsPtDeK)h6#Y< z#CZn8wJ&-b9f{v#xVCrKl6d)JWL_bki0wokbD*#(x_-gm`C{&p)Wh5ovgMu@q@932 zYnj*_aj`M>LwS#fwCxyo5sn9=5b?;5Ujp(iKGLFt+8U)5&!aetaf15vA<^`y7hxWK8F?@BdBe(yN$Qu^E3xpash(Ji+VUy848@4|sC2Ro_ zCPD9Qc{x$C0oN|*vIYFY$ucn<*P4P298!y^wN)M^pQoaro*EE}?@rL}R>O=+H#Coa zdr}}$BU(0La+vP7=Z8UlSu5wUuIO7-W@$>x!ZxMy`w!S3xu8mzgB z(D{JBeDFb45${x|>M2RGqNo4=zM8lar;x!lMEWjXRl}Q@6mzHSSxLbJ1_T%Rxwumi z>p0b_Q`BSsU6FbD?1{}CpB%-xE6DT?ne~|_q_%kw0=Vn4isW&)ihuU#Bx|GnCa0qW zIId-$YDC(qmR2Db_qe{1;^VIXQCRqK_i3z!5f)e?sc6jUpq&^B_yqWx_x}YKBt-6@ z==&@wn|YcB>jfSD{tg8$wQRHKjr>(i>9Ia}YDzNNQEo6c%~G+nuiPzkP67mM)vUXU zRHeLoNDQi{{w*EnRdc*uQx-I%ys&26Q5pY=Rhune_YKsa!a{j<5Qz>cbp(y3s%+g@ zsl`P@TanA~m;t^S;Q@!eGG^`eIsRSFfz5@MuO&{s>VdXCfXRwH#<%=z(?~z;X&922i!xILa)kf z@QJvP_HJ=sM;#8H1ju+jsFPVaPeH~{k6~8V|GfSj-a*yj$5JiLGXy*4u|Ct-cK2g_ z89R=bFgi+@TR{%6rOwZG`rP!URHtflc=xq!KuXPh^o==UqR4gsXz8F@D|3asN5KY$ z0LF`wiLjm!Z(+ni^Q#=Ph9`$qzZsH7z5<#anVfdm{DN~ghAPO1EwUU!HN(~7K`GHzcodFI?9=wq`ejd9Og+Pc(}g|>O}H2bsqm; zoot2E>IMTfSB{?*>Py$I-M675jLzAt$}DvaqY{5&M#u@enRq|G-pA#vGr#x^S*Q}a zb7{z?SQ28ZU;GP;1ag!%)qjJVvd)lNIRZ@h;pC;RE zSRz4;GpZl{Ouo>Vq88q76JA@U^BHyI5bV`F60W8!)PpL`?AT{~3xnXJ+!&8D(1GHc z4GWpb7#6eM25{LqwtP!A?%O6quc_3MQN^wegi0sW;QH9H=mB9uV#P78AX;YzK351a}&J<#9cJ?yR^(Qcszq9EY_z89;W;K$nMhMEwEaIOw+V(l(1oJ|vF zS9~B#E0ny5rUAkM&x9lOcO!KS;45qwU?(NJ{eJFUEKqZvx2aQmc{B%CG z`I+Kb9lj~r_$+SvhD_&`t{9W_32b-|Cc^Z!#KuyYl%&LhUCBe0u{>;^+pIXvm>N;D&w_%nA)kX2lAh%x_RB4k zRx==>Iw}7!<_`CKce2M{20poeQ4=Ts*;3j{-}V3tsdO?T)0*>Lk;W-}MEnEvOL`Kj z5Kb|EnpB?&-j_;=MXxr3>K+Q_VrseXC4 z8WF9Th&OJU`|joX)Tss7WEM(Nwni4>LRV_ z7bErQ*fXF!m!xI{ep};hhcW#yjSxS!la(HWZLih~UpnJgF>1C@>$nB~p8Ep7={|&i zcI{K*4Ahhs0Dnhi`q+zCDjwsxV=+9WG$Ja@Hq3m7G|U!xj9dRrabIN6&5QdjDJBBe zgwIdMm-B~zPKS}5@b&_A`d^Y4Q8AypdB{1qHQw_30{6x|e4NJ2lPU;2B7kF~WYu;w zh^#I|%l4vqr7if0^kUCrgjG_<-C}&D)3#}hyFHD2Fi%q7*Mqhw#YCs_GC~mK1DoKF z($S3}Q84CNU3K3xpiafRNq?#X{bcHIl*yvEWdPsag@zkdpugZjUs$D`{G4RGMsQnuV{qjer@Y$70c2+>x!or1nz$ zkt-@**{n4g{C|5_`k1hn&s&P8E<2O8%pym3wyBSksQ2B~IIAdTq00r^tj?t_ z+djY*>LiFZYpui>-GeJVMfv{m4y;dfT-XBtp1y~Z|8fst6<`dLUB`#!mJ@d;DM|2-`}Z$FCD{F=4ig}AZMOcIyS}xfzI4o(@M(EFi@!lI zS!W~A7uc1CwTmI<1chBilYu!8!HUZC8kDT1id{%%0jiWOYNrJU-iD8{${W@XB9MuH zD}?Bldhl8OI^Kpjgi$r=L*-RUAJmZs8!pA@FGH~tR)uz!#fq$c*AKut=m_HD%$&b) z>R7<9hU=@F@fesXJ61Ny4u>of(Hf(?vz8mu@aZkw<9_Dv>3rJL+7#W;alV*HU{vL5 zb*E}Rm3+J=1XCFtESiL5K{%!z=EbBWd>HE$ANrB~^u`(?JBNxg>%g~C~H zsP=C(On6A2r4hX%$FhgF2E{!*07-jGKlVxl=vDRGtIRk(;xx_uB>ZHp3lBRvsJ=R3 zKs1;U{k_1uL;0CS-;9nec=}f$Syb+lgOzhMCFG}dZPdC!tkR*{m#S{Y7V5xRE_GiT zI%!%Sr^gdG2p3sSDf}D3g-~x&3Bitp`S}`l6ky9aBcLEf1?_x`D-^B43!!~wow3w7 z!9hq})|55wHT_dV9WzVa4Su+5{ZkCPO3r6Z+`;2E$zj4U^m2dsxkmqKZW({&%hCi1 zaI|Je+vTPxKf{!&mf6V{A~zlBE?>bpPlk(GDuJu`ZOX+aEa%unGWsrZCcxS#D`->r zo70n=1SugIVy_Rgj9kc4J@TH*UgqEEg>}^GqE|`}g{e)3H$Bh8k6_n|PjvMw6y&70 zoJmEOpwy5tTRl@wLW1bVUr)#)TsoUMk?DABkJE&P(Q51@W(sZ-m z@xnV}*mHC`NVCcDI$f_Fj~RHvuy!_tgu&2Ys!5`Y)1t`Cd~h?iUf*4QaxQ_HmH-bC zZo)OK*J=`3u92_w<9&iRc~|6MV_3UrPpXS z3eiWC7xGL#%>=%O`PY}T$^T*^5~-m^Zd;d%$ml14_*y8WciSqku;n2RC5+K-gMwS~ zTN5Ij%(7s_-QJ{|6JukUr-X<~Nz`cI_74UrP z_H70_NT7dMDT@NWJM^0INL=>updzh&tGIkP0*E?J=KOcj(Hnpg_Vb54KX4nU^p>JM zw;D2}BvZAIvGNCRZV^_|>ye0Pyj2$VRpOjdeu%W=o;9Tqt&IQAojuSeQW$OINtdo7 zkF3>47Nqd?sjc51k*7Yvh_cSSB&AiQhBnhz^pP=IZrnVL0lm2VC^#v<>hV31`Cr+U zb1EVsTIp4S&B>$nbqX)>zFS~+(4bqtQ2N~}GoX8%jJHm5H#}?;o)2F4s` z&MZ4!>PHVvE3JmzO-KUv)W!M+p@^GEcgfwdKGA;c+)FjdtSOOj_)YCj6= zoi{Z%hNH@s*M*ZAKHD$e3JO*v)?pJ(kEjOyvPzF(n8iGG)`P&Z(J+Rn9u(!#Oe$t0BQ|mE4fz{hBb(!O z?kC6GwjDgDP(iD#;gvL^*(k5cP$Xif+c+p%IT~STd%J&Lc3yJLHJO$_5%pN|mzBUg zdfR-=$3@oi6fTlslw%GAmE)sH4@vQ!7o&q|bVD6pDf8W_okz%tepc(wOCKGnr>z$k z&q4F|Qv9fA=CZI=*C<7dzIZ*+wt)WvAOVakv1UXt2jF+%6qJg6;9kJ04-Q}WfLk?* zdG!SsR+01ec413P*$XgOQW0kk_Ti1e%YYnC@kFTVsPCZ0G~dh zG#O|4V(X>%auy!kf(9y=65nQ=Od*;+BCeVDeC#U z7?Oj}!vIRu0(c)~VAqbKn-gvp66Ad}^2i)rqxApH5nztOBA-#S*VWsPtZz~D${f=GqOKCmDqnxoi1CXUhA7Lv`FhA-C+8HI7?R@Z$y0)@_oOUaY3c~7RQ^d7ATV}^SpYKvi7TcA z1ZMPpyCZwsvd?`+xz3oG#TQSv-?iWeJJ0G}$>0`K3h9kD7jSmrI}XHm3rS@89raS!io5Jy~%|4cWEdoBR>< z(Kz2v!S{Ni96#Nt+?L*@EQ9{qPS0K>0#SpyMH08JvI~G06->BV|I6>tcsC;Qah0}c z<2oI)_{?xY&jqZ^QVf(Xo6L918@vI(_gO`GHLJYtRD$QxSsY%f})HBJ;p`@&?9cIF{**q;xpM!}V z(yOoSVEh_0F_L6iR(Y*Nq!!aB?_vOAAiVXkbUF$Yg}C>e0w2Gc#tAM(Hg)?^zb$28 z8jPt$sW8nw>8NrC0Y|$k>ooOacfAQ)xF)ASnuwix7N~sI&n+>Pl?*uOE-d~@y?Ah* z&p%VE>n8}O{lOwiwQ!mXuUQS7kEIY@tF=FKRbvYveJsIx z4E{*&A$9H8|DCNCH;q!I;fVRB(3NtbZ=eo>`4K+N^UEi++5N7uVuJ>}u&9EGH^(*w zXrEc(-pohXo(?!lU2q+)3sf1lQ|~y+LNjFJOk|d8JeXLliv6+my(;+^A?maq|%aJzswlX>6kw|7~r z$c-hwY{9!SwW6WGkzjTT9Gc`=I(DHSP)BR|*ZQe8SdCi*H5vP`gcL@$78Wa^)liHh z2q-^{(pl1fC|Fkz*97O3$o0M}SDFA^6pZ|6lK#MRMo8<_vgn>|PN9X}CxXvqUg3R3 zr06A`giKd(CJh9@m?;Fzp5z3*Db21!weZKohvHLkWeS5UI{a!mh$}u!#htj)Xw}tw zYHaRL^@t?ltwxrlgDA9?HhFh#-TMemf3`-#8UA&IgJ-zlDpXNyK|&rz^v}#9UpS9a zBrAM;xjBxgaw9C-*GRcNPtf8fBV#(IAe)%Z^DKFNho#sur;|aIq8%7aovKdl98CF$ z-TuRUc_$5ljBK9wK2jbQ^DIDODHO56dVeH9e0wE}aV80^9IA!EE4YW7o(%szsjj^v zKlw&}we*iqK?jB*6HTEtH09Ue)cV&>QkKvbn2({6t5O2T(^JR-OtAXrk+{qQjPuHz zu6Cm>>Ho!e5byTtBIt9^2KX~3+)E&YZm@%4?ZOwnF0~K@v3a#rm30WOPZPGHFr|$0 zx)Ed#kDwa_Hhu&sZ2K`d{vCKr(CRSuXM$f3*16w+e=<1H7|zs2h4?A#ZQj5&16#!y zm*?TU6lipmjZ(ZKs{;rlF~m1_x5V?6c64yoX*I^h!8FjQ-ztA$vvcspd55Av+-7IC z2U+WUi68SHI|kN3TXPhH&hza%js=mgPT2@oj}Ql&Y@tnSjPF-B5Gl+cq)+n z1FW=k1x5d(h=VOKLpsrAZK6YXJdzMJL%gnGjeP^FG62nhQfEa&cLR?A@Aofo=bazf zQb+Xl1C=j1!>vN8pYCn%lOk5AippsO3llIb9YBpeu6>u8Uc3@jWeWdmlPP@H8*?b* z*YT0U94s+p)|s~Bo;qAIImWbu@&7vaamUAk?%+F8f8t z!A{X>#i@97M1ay+gU?2}%hgf&y=affnxKOEmvP9nwG33t{U-z3#V`z%Jv7np#< z5K~n(2q^BB(Xb@0469cWj}|4I9w5s;;zZ zN58u@AmWbV9cJQ^v%)KPq*d$f6LeeA$wEDlKrjz`I=kj1CkAae`iC{J)FF;c#7O(m z)evHSD35&^a-y}Z3_VgS63HM;4-jlugv!z1~e5*mXOAszwWHL%{ zT8$rdSqfN6SP4SJ+rZS-IXo1~-9#|DqO<^gzSHfsT$MWvlv4c;5qi@PZm)UkCr*ov@^O_r7r_eUdyIbG*D#_|KGz7 zGxaV3pVRhTeBn_;YPkUUteUKC9daOk*r9(PyDYeB=s{mG0KN}5B8wIm|NBiNY16iQ z#76`R(3;5rdcsVP3WSUw`tb+l>+IYJp@|g9-B>kXo@+vnHdf=bq2Y)>pTlpe;N>Ru zu#kq7Xtq%ziT1n_&g063a4j2@*c938l3ptHcTk(^{mm6}dZW2@(i0b#EkouP!<$CD zd+vK45qpQ#(#fqR6Au^(Mh7nvq2MHkqW-u_fsL2B&gRqg-KvGJgBF5bfTd9g0Yf=9 z6rJubK!!*QZ6^DY!@M8x#Ghus&bJ0R4E%LN66w2gX|y!*jnxUbvCJmjuR>r{2mml7 zksB!d@iZEG_Bq-Zb9qU%13n*EiQmeSmcD{sEqoWBPG)SJ|mH?*E;x&;%x^{!CgA zE9pa7lPy#OIW?bRcbyc2Aie6k3c7yNORu?;OJ!n9U;f-V{7^-a@VVrHRlb;sPqV-3 zsF@`$?@#hGDBz46eQSel(>0A{CZ4Ww0B>b_?*id#W*MfcH2Y1ptcK`~2L1$0oqqUB ziUD@ZsXUt;1@^fzxR|kWiS(R1p4zPKd!@dPk!rG$U_Lti&CgPNh zx$@Q=H#?qENsYjvSMvM!iVmw)u{}w*KJFO$pkWMP$|wtr*|V)(@DHk3v57!vSE8WT zj7S39a7$NGYV&*$d|l`F+=70ACUTOhg;>A69&*`aphFaei{%6AL^KqJ|F&B3TRt~; z9o`uu<`_=~PgkjWRIK?qnRgT-fPbEI=&CtnM|~D8h_SQ0Ay#esID1=zO|5dW2yhUg zr$@Ke9AYEk-Z%MX1RfvsKp+Vlg`ey-pEQ5|20qpMx(fF#ZNitc>tL|Q`lt<2?iZ{T z2kp6-eKG5Z(1L#x1`5;4FWi;GsV`*fb-lVBnL*$%?)OqM0y`__f_4EswS2S_&8^RY zoet1xrVeUVAIlLDHHq|x%??<8SolZAZOh<(zdD{S+lMkQvJJU<*t%d3J(vui4}Y-X zJxGuWcngFu(zd#-0{!$7UY>h?jN72D`o#)m8A^bX9W||<%uA-W8%+rXv;65mH5-+< zMk9`t6m`KC_{$L`cXc3+gtK^4aYQ0t5(=~Tmk|_!;KBR@Da=!KXL6ERuQFGzb{*{n ztND*YmTzSjNo+rrsY0Vmi)i@oZ{N-bw2$?>MIjkK13%lY?f|L zjuiEr$XT{uxs!FeVu>QkZs25cEe zdj=eeHQ>n1qO?nVWy$BN6y0y;Bh-GUEq{@0x*Aqa)4bcCl0Lai(Qt$Kfz4 z{+Z}UgZ-hnud|Ou+gg-3 zPQVuQaG25h1~TIcoFw%x;{2-fKbm)Q-}XK=ej8@qk7QIyH`mhavNyrkCj}Z-rVu2Z z92P4P@6`4CN?Ij!Oo*7-ts_qt>r)NUDJU?`irUFRqyPhQTM#&@Ep>39V1+OIqVZv@ ztyFspgB4`4VVnfEp>Vi$aegp1!%z49vnZn>)a^9IWG-<5#5_5T)iba9Qy7+GAp=Dk zZmYyHL1(WzLgJ;LX=Zj@XlO)+N?7d_LCs_Ex8jyHo2d57LAor8uCC@MGnz1K--8CC z)*7!jK~$w-C1WN>6;BMt5aQG`7qG7e6L6FhsIkSO(6W%qkJ1cUf#u-NIKHjDD6!zI zqBFd*Cnomw-y(3BHs=76g?eQ>q!~Alv$89I#ze39zepNrFC{G$av!x%(}m!Q<+GN5 z5PH;$c*x1#Vy>yPc2nXZT=mFqF~kNeF!ipc`qAy6YDJi*(x28^)SRD0$g(oJU5dJ+ zfyf{yE%;d5xhnt8j$BfJ8P#gF(y-|u?{v}GUeP`QO@MFumId{Hcml5rJ{RJL)Q<`6 zW7>SxU9UNM%^Wys@(J%93=g4=d_Owp_jfeJneU1tvRy0W+5LEGdVUYCRGJD|b(1Z3 zooFoFfwyWvFM_MQ{~S@r2VW@AhV4(r17smDI6Ib@dVosLN|GDBz)dn<3aM~F&+20f zDw&viumM%Js%m4%dkY%v(O>lQ19Cgg3P)g*{g5C(c%ATbUd0$w%jV~I0CvX~Ad=mC(=0hT&K~qxXg{ zZ#K$K^Qx1i2iFoPZ}c^uX<7T;6;tEaFysmPUeIK4ayxu%M2uksGTB;DnRTGi*nka1 zjdR5uvs@7qL~FDlhRRo$LLth?4@d4_S(M)#+bpB4Y5R5+hh51*d`j&VTT!=bpH*Kt;LdAtX&uqwyN|%LK#T>yAe6aNqxlPLXc}JDj~j5$$TD z52i~b?yq_S_fdbZb#;-C{O3InSmR*G;A>WGcf`67q}@hQb=HmYcj`i1G3BRR`Or9h zwe{M76#!GFf+m4n1%9br9x{ zVpP1cv8j^S_gR+)WhRR6QBWEi5kWYoZZHmfss5YM{BvGZ3a_|~F6h7F3t>_!5WbZz zr?wCHX7%JWiylZ*MXq1%fnh*ENJ-GY!t*N@z~v!Gl5v%K7gQ3OuDq?7CC`6vN~?Mb zKt3dNBATv*)(gInN(bGcp;EOEw4%`IoLcdm+bSGI^EUnUX;&nM1LN1}gUut?X*{U9 z^(W9y`_^2o8`1mjK@a~ORfra@jMi{Vn{K!i?E{T$#~K$AVdEJ@;rAXvYoL6)ij%Or zgWhA92IN$Ui~`-FOWz8Kut4@g&8HygxmgjG78D~kT_yqf+V)6AOn^kXW-t>vd8rpzJjbT1)X~Vd<)%b4ZNDO?T=+S`XEO9Nl zm6^JE;mf=6c}ThqVK?>f@kp};3Sj)TbZJ0;%9y1BTdZ5Yt8!%f(r9y0_M{I0<;;{% zHh}A~V`&*qv7V+-=#vZGd(Hj~%^IDzGnA!=m`E2+6C9A=wfuKyk1ze~IWV`K0zoYR z^=4c3wgGxGp;Hn%ksW+p;&%#9zcF^HH<3r!!G{w^0BJ= z5eh%)8MlZJcTIdO>tDNL$Za$w?bX;Y$wg2W|H=7v&q{I*wfT25a{uqpIQdr=0g;xs z@f1wB;~>05&fD$a(6fC>k>SWZ;-WehTW&m5j5>JSAN(n1e@hF8N8S&QB?~h$>MgE5o<02-cR=@PR;zCNssVE{?I!*`d+pt#~ zAZ1Ye`iVp$hTo~Oo$(W`iPqg`UqC6O5~&mi7r(Q59jQLLn)PVtX)M2)#0%#R4kxj) z;9Qot;R@Hyx!fRICFQ6&G!r=dAzXBf0?fbIUf32k*Pe8Rl~S|y(Y|g8A{)2;6rUO8 zrnvIh@(fqf7>EE?;&d&A^gTLpA$FYf+J&0@;LN_ApHm_UqB*$%Tj=ZvvHIF`NbfsL z^RbJ!I(#3+-)WQ&1gz3d&Y6imUssAUehL>Oivb$;t5eeU33M2X%7k!X4$$Y1GI&1) zZCy(D_Y$mc%XtXjOzj-03J^+Abv7+mw ziiLhwZ^Z>D2DGxQ?!1f0A@#W>V-D#YQ>_$$9!rHmj3%O2bL$@5Npi`N&|RCqbSCyM zhcZ^<7KDy=_Z*u-^ULcCa2m6g31`NjU5y6m=J%p?gaRWx~3CZ?K0MG)!XD<0&|1N{#PcQ;rBlsIX`dK=twKa zN;)^#_KT*O<$2H%+d)^DL0Tef+nuBc08^G`>4Bt z)GxbLq=~R)S!qw&H)iI%E%-BM_n2 zX3#ZE$$%5ER6qY0NKN%lpkJK-)b5owOJ6wAMnsBockG+nVFa?`)Pm*Toz z&wf+18J*Pu7PKz5j;X|4lH$%= z0Ewq{9}Z!#jOqk?>qXp1h0X=abPJUVWrMn6`yI?%%tgI-#k zsm{m6hFhG(!*$1P)}nGkIVSsV`tP+t^!`W`W<^A?P4HxKe1jbfDia{ht_kjqZJ&_& z`;WnC_hI!bl5a@&sK(zRLGJPmDprVwV(+CI$DFBa_WHa_gJ zILkr+0=8z9zV%uwW?zRTB^C7x+6FqBZ#>V_4_I3Mm;q7Vd#n;B9|h*D2sk_~KHSOD zu^nrRF!nwe-cMz_knWvcGUOPC5Loq}SYDtGYjB`MC$NBG`!-wDz3DVy6nv%a&*_UcuNZLDy)eY>uZ%5NL1JE2JC=7T$syG zCCATzb;nj`!-j#fN`Qy0l{CS2#)r1x8hiYC#Jo70p76Hed}6!kzGX4jzZxEufBa(E z@Z0*yB(#wKx{{Z(CmwR|47(`NfGeO4ckVz1-33Nb-9MPOt2zWC@Il?P)a$Fb)SLKmlt>Qg4H=wEH<(#mzR3T#r=hoaDDPrQ56YM! zfIzw_uM5}|Vw@)63%|vOJ-i{NT!bSzfI@N{(}iGrKJ}HxmL_L_gDs2%4%f8XL2Ecx zjK3^A!otWOZGc*du&IHv=K*;SE(T|$T79LRyAv_%jdK_eYK73F)jtYQeGr~@3yk`< zA%|O#^d>qLfows&hRwkTK@eY4f+pF|O!r_}0_w;)!xW$Ck$GdbIRu1&+ZYQ zxb*{Uw0|(_9Y!(WcAk!q>VdtD|Fxg=+;P_SI2wQkF{%lI2MTYKyHBm{p1Wegg!=ISnX!S}0_O&UQ(rx zjYLE9=4DP59rKedbNPNE%PPLw_Pk0(n$c8#or|}b(r1u|BcaCOr%3-$3n&!D_yqf$NaMw`@Op;HICgWOoJf8cvXzW;xY8*{=g0@#P`;KNPg)Obb_Goo``` z(m-J3w1gu*gi2@C3H**K>{^RM^LqzlKqg|H0>GpqIYd=vNVku~=N)30{`}RbCA8Oc zC3ZW?*F2mNjSOfK)wOJ8Gk6G8q+vlo_A*Yy@(!j{zAN}3C(vUMU(%vrobq%NWAGiP zD-`&bCw4LxXOx@?9qTCkJJZ8)3!TR~SqV1t-ALx(*beEi)F-2e{l>FK-Vd=6W`lgCnE z&j_Qb5yyB=nA9axarLLCDyQ#? zb|{+}Jr(i*lWK4eD_X1$Gd>_+<>8s5>(-^J*;~49Ei*Ia$LLE~OBFH@g2QR`0I%F` z>*3TkawTPQAH+krLT{q)`CIn4bxFJ}GG{3!^$hA0yy>nfmyZA`k*$%{2TMl=uXyqw zv<}=9$_jF7B{7@U){SiS(u$l{0*%Ad(qGOm2KIcN6* zeFp5rQd{X#N8D{OFK*pbWD{+kR^mGOH)|43N zAD5LKdKHJ1p1c<*>19P(W6#Cx9Z_N?eF&0z^bBpju4h>rD z=VUo}7<_(PKUZI4BA3NaTXqMYf)$XQO2M#5X-3?pY_jmT_f+~svjw^4w9ddXsqbwh zXY~`*WbJE6px&3h~Dl~(BV4Cg>$-^^&Ox}Z@E{x(tIND7Jb9noPyU%Pza)i4BiJ2uK zx#0KL0-ru;R(bsuHzE-9XGf3tpe=y~UdM~Oew7ZpJ(2T^s+);i!OL)0?K21UgPUN) z+dG1tXcBxT=0`<=e0cjJNL^B0k|J9YTIEg_s)`VJQ0!s~?7L7}oF-+m1NLN8WarNC z)ivtdA%A|bIc*<%2l_~7+!#$r?+)VRov~a%*w&Dz`f5HZ75_>-f>eguhc=r9<7{}M zsO$$=$g`d1vRa2G2U1(5+6vpYzWD%H#ZyO%9MY{qd1DQ*e>q^X=CwA!$M#*(>RoAj z`J~246Isp({|P|iC6NPPcT@Oz;q4Uh8fl}beyNS*ojdVb%1y!>t{^9f**`s|bQ-J` z_w^6<%SATkS!Ea+-q`yW!t@zjxbEG#S*)3 zM5}jL#bmu59&7~pGN>8=WUxZKa01};e~{zRS>!yanrzc^Ld8)?R|ji}TS$Gf5OI#j z6sk-yZhafJf@j{_MH#=U`d4Z}R5!(ZILGXhN8w^FEMRjV08>D$zr+qgvM}c9C{nn7 zj@V5CjwIMtmni|8yo=EJ(lF$o0@pJhp4hW5DIL2DV%*s=NbOcRaG!cfb@S#ku;F75 z**c6d-oZ+0VG#4B{JOE6Gr;N`+iwspp$EWNf<`VGU;Q=>S6#kMB`yB#yzI@fxs%Q4 z8VP5Edh{=N1lURpw(Ufkq0qJP6Zi#;!i{unc>)VH-4+4WJ@)UyUw0blZ3oqTRB^%z ztw3NkAoZD1ky0Z9lwLSbwy9PrdxAeJ9O_il06Knkm{-J8 zFxr|XjI*&Q^m0e4Z5+)=PqnnyN%Q92rs;$WTki1~h^yIvKnx6lg0z8N$H$mVH7lbO z@<|_0r*u$q^#B3dzyZZZ=k$gc$nFKv>Qc>CFVX1n?l4R*Bp188BSARAJ++$vk3*3J3Ue{m%`shos zT=e4-fsufVgfXVXR!j)+K%GH=nxl`To3y(MP=+t;g}^oGg+WXp8DOg2wCiKn4_8aI zA82xN0{dkE>3De4@xOw9U|hyE`q%7Bj_xR0s`3+1G;C|LaiyxSV!sShRNn+2IJZpL zla`F*gh7p1JtUOM4x_FrzI$DhwF+ z)HYHLLCbkok}?|!86`825g`m35y7`&xM1Ui8~s0vqQaz%MWwJg^~ zW0ZN&JIjlB*6S;TgvD+i8zzbtrcj^TqL25YX2L@SbG~f14WIQ~taH&ccs`sS=qtS4 zuqd(g?3!Wu#a{*@v;^lwzq#SK>{b5!ER1kWrCkW^%Qh}mj0B_k+Fm&c0)vUHBDLE>3>c=Q$VCBf|rq zJ(U(`&_34t`}c<6GJNaaWbZOi6p070WuE+%>V;Ea%y*s{A6!&W;^|7*q#Z?SgB4y6 z7^c?E=C|hep8b>9o;|8nyxFoY)7MzJt*NJRFc$7d)N0SoW&R8aT9t;f4>9XM*0MG) z?#4!r4yP`<4Q-gWrmvU;PwbQ%e8Q~zM0c*5Bqy)OPsGZ^4@^PgE0lTWauIhI}m=Qa+lYv`Rv<=EgV+b>g< z?8%~?pt(I8-kk0vL8XT#0>kvL3pJXPN^hE$H#`^jOZ)%$TKLvAC0ox_G3PL%NpjgM z$|I`)SBizqB);bp*RUB=&8VKrD|#d#e%NrXBVcG{=p%Q|_GsM;JnXN1h?3`lU>`0e z@j;^mf2uklFTmT=qwQg&{S0OR`)ko(P|iBY=PO1V&5<6X$`O^G;6g1S3vE04hdPQb z<<31&X6%J)%RCimiiZdIW-b;V_IjSIuLYwZ#b>@Qod?MX0+t3JCm0ED!hMQGQbiyX zVe#0tZvMPjX-?XLB4K(^rWn%Ci;PgP0IXbmj(O4DF&Y`Z#s6o|Ve&yig<-`UZmz!- z%Iy0w3q^zC&rs1HqoHK{dpQu*%hw=uhe&3l0Z?sD6&2qu9X!12hfE_OTL`SV^8Alg zRi#l$7i$%sc#GQ>6kmO1>@&tA2%=2TX%nh3_ru%8$ zCfaVEHDMFSn~gSiMX_kpovGvHkwR$W|Bn%PlblG(`xT(-1A`P9)7eRy{dSjSE;wn1h{)kSXv#vwA{ zg}wuioX|HStUIy>$ErFmcs?rV=MeuNle6w;+_J1Atb&5&*Q+;Fb4uY|FCAu1Ln070 zypfge@xmGdl~-ygO;^0tFDW{#c4RRY$MbHh^~*(h?976*jJ0lp`Cu&M-8A!UM;neK z)0)efLOWB_7P$@uhQkFte4{r9DT75D_{gBodBTy9`$QHNyTyZWM=-N=3w92HveQo$ zLmeUo2usXw-kZj&Ic+_$!OO1D@LJg~bxA5f_`?dpvOKc$wCGRlbALQKpG-1^j~Q}aR!Z`FHDvO3D)q{;1f6->(Jac z)$)Y6@bDNhAan=xi&+I1K%IcB4rhz(%&y$4)*ly1QrEaXW_D|R%!TngckB^)dxF{L zkw5ht8itj-<z~sw)UP1+E{M6$@ zhZuxt8IU@V;|C$RoyIe{$2!ZK`dzN~uB-LS9=A!(uflW@JreK3Ai9&@y28d~$}}-Y z=;%e}70Y&j_h5goWYp2KE)v-DqnKD?i{wP%ST)!ZVC>z%pz`4{S5pnq%xlKkUmY_{ z^45fZbBfUx>2d)q56QX3lQ$sVWIH_s9eSWe1R}>&!eKf*y6K0m)+)o)|La!8pZ+sv zI^0-RAx4>XlP&>=4?MDbL75!ciJLe>lM_hsLqnVoc$gTHq!QIYPUYQT z3|c|aIgDchgAF59jWT&Y46mxZ?N_y_;g5c@l%2hh0kY?;$Ce8r!RRR6m>r;9=|PWq z+Lmp_JUa;v7ExZ~+)0ekk)2W!8pJ%I22PY3^2c6wqyK(2aF!IYvBo3I&a!&jL*gFP zPHlYOW=_Qz%A;ld*F&?Kl?6V#N#N9;v`MIXYe)34HS*wJjNZPm{^|@+7852~ z4L#~r<0M(Z5>fRtAM8^uN}O zD8m+YccMD&8;PHpSomD5UUBE;G;NlU!0ABWm_Rzx{(+cZ7PiRL2}{I-W9<_T=?1=d zpkC`=mLTkvphV7U626aPPez_B<4@Dyp%R*rdcC}#oD@UKah0hk4gdiiPvnlABk~Bj z2K#cRGjqL|;9+%i9~WVQz%R<;y8wcYtj|L}=kMx^qdzHKcc8f&hU~cTQuFA(!eVRwsQmL5{U(UaSRwQxP$YlL zgVAB*1_Drg;dX(k!l@~Y+XvZwSMil-^-0Lh{s#)2eLGk36EsTqUVFTUuJXYh9jIFc zOEyHQTFuOti^Cucxp+MlWijf#`xC~AC&H&Z=hFb-gHLalE-UFHdG#DP`vFS?A)C_i zcdK{2Rfp_8HjAB0A`tlPQH~VvkRTMTiTkA71@d%}-`VZ6>?;w5)uqG zlD%|Wyq1ep_k0Og14Qw((JAa_m=!knGN(HAeekBUM`8y@eOLYiTh#Hq@-vvru!`pZ z?JK9eBwNl@E9WOz3hQ5vg(SvDUaRfctE~*N^7Bumcc|mTiRrCV?8UHr{1S9hS<(02 z_&5@So9t8mqIHmX?u7D41oG*PlpyC|EdG`r3YFuX97CIFCGIE;Hz1TflcoAPDtPY2 zmVX(Hcf&Px{maG#HrZiqKfgf6AfSqqsjRvTi53XnfhwE1p>pWJ^U2I#89?N9#F`hF zWT22eBQ#c!gwS2UaccA@7cDiql=T}e$y&;hDah)WgFIp-er%%7dvwjmlw##n$~R08 zYC0i#Hj}!v#F1g4*D8j!*)1B@Pdb-s5F9gRzO8fe+;O$v`~t$>u!?qF>P+*s4Edk& z?Jvo+1$hG&z$d8l#YU(w91prlSVO;wHFdz1<0MG8nXcR^s8`HaDOXHI5aC_sXI8-$ zVxg9`&k%^BIpMHo2ff&d-$zy5a0~)qp`>H^EkVel_wvLTQ=03TBNckJ_(gGN<~R}w_(5(@AlV?F zuitec(nKxrS5i7k@4uqX{!v~76)8|J)s*)mz#2>??g>`9mdfy0lNH{n-r$b8bj!p1 zOb6lXvaKl1uTzqXf;pKP}J?(!xNE?aOmm>M|(R2zMnVHu*{h2BKhZI zsxPZ*da*6W!lIm}DM76s9gj&%Kw?0M!=2Kf2GC1>*tiO9lcwlYTHgh}ZP_hjp zuAHJ=xq&Nr1LH0-yUM{nI+Fz@GoVhVKq1bpl6H^hvYH9NvgVw)TKChM=*`J-z7~A2 zG`1OR(B9W*D-c+x)b@i1_zU`^LC|;oJ%urixgPPCWG5VYYvD7jt(YV8Rc zb(m^PD^m}19ig7tWaH7on#gbeCL42hR!FoyDtwP5^Zfx^y`aMK>@5aSvf)1W^OM(b zDn?RNr#Ml$0S$lUCcSRP!W2)#$<&5X;H)-G^p-9RkZC z2y#8+q(I_M02u%o)ldeHJx)}v!6^!Gj~&1-s))BaLfUqQsU#E<8OQo(yD$@`KF4!1a$kxp;8G>UepYojp;yHH2sg$R^?@bvOJ> zx7$+4ebR9-^Eo^AuhZ5&2Ovscx}5DQE~w;BJ?yPIaUa<@muqRBy+-097GF{xYe=j1 zs~O+x9N^ZmhNybl23snrERSY%5QQjpDGJMR+Q+LvX+OyIb6t2nH z8b`qE#ix2sTn)Ify1V!QwWU!mcQQ~T*maL0Y_LMZOH^Q?O`}ubrDZ3#=9`rGQUQ#) z=4TMFN7y*dQ6Md6*_!|f*rBGs3MjJz&gaX@mytrgPuph*#V1$D#T%9>dMA@ERATVh z241V;Acm3+ZQ3MFT%pTmtVLNA-LV_4Fbz}r2@qR7ZFH77Q!o}&B#9XayHpZ>hgcXy zf5D!w=mVj=PN#KHFcVGUqpBa+0};VZEyA+rC+UMB$E^~YGCg#x9zj)^s+Lo7a<0eE z?Z(d2J!W4Q!E(w|8{r9!+8LUBt%vMTkdH^k#L{W-&U-ytm>+!cmxmGcrEz4^l)K(- zA7cW8>2Sd*eKGUH?urKW>Pkot> zN^!7s-P%RQAzqpSR4W}&o0SS#FEZbfkPP@>8jQ(`kH|gnWx^=Gkjy*NA==fG4`w7b z_t*ojA7yXRvrJmx#+lVH_>UjPA$$-ir@_)Z^D!xu$=bqp1rEBr2Sk}z}y zrwlDj$z6`H4Mz{T89`j!#p*e(g=D(I?hgQS0jf;?prAo=s{eP3E z;FHZ{H;SrMRD(9X5>V0}#x~)h85h2@EKdgK8sD6hF2*C`w=^%UV^^{WJJiSl{C98K z`M{zWo)MXIIj6lgCkpInxI+xmRabu99dM;uTZZNvIgv6$1IA3 zMt&M#bYT0yf9DYDvS$o^P}?_aQ=`+LH57SAsam_k+4%O|^o-Lp#%fwk+01^3aZbq2dxwy^Kvbz2U4<6OPYxeraKy%N*9o= z99AOb*TSbNavskYOT$gOM2=V_3H*O6 zv!J@TeT7F}Gvl4|bndW)`v|_U3mvY>QcJ&)^V5-%r`W($A!(CRfSvZj2NqVr#>Q^k ziDn*WX2Sa4e2#YF?NZ)V5-OdL0%7M`rsF?@(ChL_-Eu#NjmdU*t)eT0*iA7Y7sJL_ zDD`+yy`B@V8Qh*I`8<*BtQLNU-r7XMWIXh{{&F%Vo=GF`EUApl)Z|=_97BNur|uTv z)-Oit($MOzDS=rF9im#y#E(7|Aquwj}XM9iKF_F8VJ!5>FgSs>_ zm}d6ktA9t%D;V7Yl-w83qLo9{s<#9$Sa=cKvJpyg-r1NckJRtIw4UX6jl7SFUMin7 z>yk+4P}%hPe(lQDB|k80fK>;rGj?yEHvmQyy=GZkJq$zkUb8o7VMq$(seSucvnLM z%&@DrfMhSJK%HidRT(J1CWR7-mjuy`08sA#gxkV)%X?Pb+%d ztPCv^fjIpX1O*eNWs480aFRf3L44k80Ii6pZN^5!&!hBg%n!)t5vtdTS5+aJOaytw z6{hA20oip%<#c4tx%YKnmcNm=(Od#7o}w`t?R6pHC8?|H)04NV{)=-9cfZF1^bQBg zG4R^yh%@Nl2gcJe$s$Ks!+bvbIRa)AQ7e=ZDf{IV0Bqj!q22gbSb~5*A6kC4c8e+P zxVixi3d|V3hJx5|+jgo(8@nkm3@`=bQ19lXleQ%g#Kd&C^xOdJCHq`bMc}*>cQ$5T z{OH=xU{tNN%V$a*n0ON-l{)=RYG)ifExzVlR)9^$xf~mnWMfY`jxmo4X%r~RfS>#v zCLbS7Hxw^r4e|!ExMjFNFZ>lgbIG>g$#k}ZwV@^xhlyE}^wmK=iv^f%cYa373T>SB+@$vPL8mvgrN~Tw~cvT*ZG{_`%1*V~@XD(MJp_;_ppC=;P#^`L`v5EnPZA>&V+Dc>GD_1itqMEgbK2^T@4hb|QJk zk8)k$PPQabYz*a!Y3;xU>Cc6aPDVj-t(J<0NLpP1#G2*ZwhtpT)<71Tqt&X9;cPJ~ zI$sY2fyVPU{i^{azWiUh7Z|&4>b~{bXqp5Z-L+y8i4vJlR0(nkqGW?)x;+#eS=EsV zr^HiH&J$0R2W+OM-*@CTvWzed48J)K?Adnm_ssgLrV1i zS9F){Q|ro(3UOdLf9~qKR(EwiujC{nr-dN}PGhj16w1?a!lf$e!ceKk z)R5pxmMQmxgcMY?3a*8h`-@Sorgt6rK%v#t*=GkcKW2i+zhV_9YBX~IdEP)oCiq|+ zha}oZZ+Tl|Wl&{xRlkm`n*5W;yb${681X?5bWPWSEJ)zx^@bhwt&CWKNR=Yus=>-Z z^?#wu?{=Nz_xb#e2fy&ZlH6x7@hCDVY}r|%J_cUd4`cTJHqP>f2ZV*UkP}cVkK0Xr z4Q^*{>~V7(>bc?-IVeEKxYn(|Dl=S6{xamO`SZ~mN>7#^CVj@W!xd6HiMQlC_U*;% z!x}4kZP7Y+lLmFvst{ytL3|f52Z{glHl}O{jC?Lv0&)_RO5s}Hd!(BSc5jaXr zM_qs#?CEtXQyiVH)Hb+}*7f5!rf6&v&hFwO7Y4Nzp#(1zQUcay3iH{UN1mO>Ee$V-|Br1fiS-4Hfn)2^txAl z&F-P}+B9hOsxWejKZcE;J0K#NGs@di(kw6GQuCsRAhbtD(=lc5tnsE#op{qzO&?Cy zO)`N{o~Nf=(JmWvb4B*iVgJ=Wj+$-0TRwFWx?>qe*Nwdz-&S`acT5Q2=b-0ygjo?0 zZKPUC)Add-X;XjMJJt*Y#H z&&E+?rrFZ~Meh0ZscD1;%vZB}I>Q)<}7)30!4{@$XbDU56{mMCH z>u%U;L2_!Ws;romL|giuDDgLR#c2f0a^ zzAe=@e0nZqS%id?O*NPcqCMzhZMSpl5vFu?0f2ya^@YR=hvIGBp%|8<>pu=Yds+RdX zJjmj|p5ESi47HZ+PSa|J)wCRO>*eR?iNI+gCcy| zpgtnC@Yqz)0Yrob%X9VH8eGk(eD)2$AjN=3ZY@&e>Gi<88*I2kuX(#bmJ>Jy^x7^< zMbxpaxvTghKW?cAd<@9oM^FV=IssDZ-9n=)?HAG3FSBcL>2ZdA8P|)|gT-zs&~3T6 z=2`#i92j-Ve|Gwvu-pR@Dv{3i5h6;2A&bz&Gb@}dya?D(pQo2(I&7o1Eqf&f&p ze2VTQ|FRc!uEot*?Cc?xvm<;*`0~M$5X#B)RB~=YcOs{8S)GN1UHg)!^CU9^sP0+gp%;$i!N~$1m7fin7H2xEMU6!yz4J zTR^}-iy{DQs4mN_vbE)U%0$Q&?;e^~@_lIA@k~lVWQ;~sP1C`tUA5D!iQOYIX8)^v zU*WgPt@g~apL{+w$^+b_F8v z#Kx*4tl7u#s0%-)BJ5bPup)d?m`O&tSg>`LAMs2h zvP-kEl}024J_zL$MjIWO@7fvd9WdX;f?J4rpqd0x5QRB66w>_dWzL28HP)N02b+?G zO&6bUMkCtJBGO0&>?PS(M{?#THyRAs5xGZ9(o3;gk%KD&1JP(z6+b)7$WWF!&qN9^ zBE$aS5&SKABeil@FL*RWz#*J}3dIa8Do0Q0P>b@L-=oHY(cqUcrV+}?7-V8=GN>^SIiD_-w! zh6=)KQaDwB7lfzj-HrW4b9;Der0z0B+8659IsMb83F0>goi__Yo?8$V9DY7BInn3@ z@B(ipYNcAB5s08e8==RMB9w`SWQ@7QQ#R$`g@=N3oiH3C=daI@`a zqtiM}?y|l|YG)#&M83B^C{}|AQ_!Iv>_@%M58=K26@rihL-ms8W!v$nBkG#bRB+zr zz;FOAr4nDiVgWyZ8*5-`MONmtucKQw-ZeWWr(F@qcw#0G+H(T^us-}+mWCL$!_Ly( zMNOd)|9oO7)|wk(gzM{!UNsmZNCLUMI%Y(Jp{{WUV z6YYzb6AY6`3|UqVT>HNl&eK6)o+9AbDYx zu0Gh{az__?bV?JsntW0lHty5+MZqb6PdG{2hQ|1-^Ne`Xc!}C7OfLkX2wLG4jd(dZ zvV5r>$E%YwH(OQH+*2YaZ^I&oXeV26UB)|w#no`(puZ>QOc99^m!3G(=93o?dQPn` zL5ZasokEs5`N(YQcRsvq$ZrgLJf)E>W1~p=$};$U{EkSg3i{aqHFeoLtz9U|K)uu@ zZv0$O=duKm#Qad>(IK-=U77?sp7?n~jP#FS>oykp&L-Sf%j$X!t**1)o9Nlrq@0*l zm7!_O4G%Y}8!9^K<=Na8kM+290w{Mx;mAJQO^Kz>yIxaGU_j_t*=!=N71@QRImd6u zJNmvoUSyRHD-Q7yG%y*heS~4{1b}T*Z-p`(y%;0J%TzIOKC`=Qv0#-OAA6R3w~6*V%Ff zXk^ZOO!=HE9?P9lVv`q*enlD^Kge$9S&^E7Jh{udA(<;rc8u{82w`rb_SfVWpZr?X zc!DL+#pj(s=hp_3Q<`Nb1=FaRma%+e>WOAua=>e zc}+oz1PLm&x7h=GGpnhv_P)?;3VTKSm5Zt&xM1X~6mt-sL6-@92UTH&to)2~zyc|# zp64^%=#yD)lplUKC{jQ<#DMSlQr$VRnQpY>|Nqw)+=j_@;#5pV^ty9s21YFmYInm5Wq{~T4a4+H-9?YW4U8n{%=a;7#8|9v#ugVK0oX?VslA`P3j4O zV;$o8O%$w#Ks7Hh>wruM(N0tBXW`2?e!l`26&uykt0hEIfVj@IdLA)J4I1Ofb zAmm?Os%n6_e)2CIiWtGz2$<`Q&vFby?@as(!nSUmN&AtCANU|?&s5QSPC}3FAU1Y+ zFgMpG6pP^^kpuV2?_%KquD)4`$N+d7!-PL(j8uXcxSwpMN)?ruBUI_kooh$0Qbe7T z#b-4Qcv3jlx0b>}waml@n}>7@K=3<@Yce_O>}BZ&4y0`=6fx_++cZW1ZIMQ{-qICB zUU?V?0jq_GYyCvFXW|XQVuC<9o7{u{FSH9IvlV8Rt3=0$++29-G{O8}s8Uht=kC#} z#%;A)6JItCf>kBJ>T@WpUu<#Y9w{}b^=dJ{ z&VxI(x0|!1R`p6$nDKW4MhL66gBZJgTd-o-sd+_;;1`Bd}orVytN z9Opc}jkm5X+78eO34*KY`U;bwOH4@IwG%bd{-a`e$V62+G@QN6Ehp$W2>p%*C}1j6 zz5T|ynGg(D&Gpz*$P65A?0MJbP9XB&xSz(+JN>q=V+c^oa6+1aHPIq*X+vEDj*KtO z?AuGMdfKFRJptdY+Ab6!e!QR<1CoGN^6s6eS+{f`SSjVD)JoHQ;W^`G>bG^Ls#pab z`2w{Xd+Na$`=6!d$_HPO#EUR;=JbwI73#HpslMHi#^9H&f+~!1xvzJzQzWAMGt`L& zd0>ZOSF&?&6`y*+KX!R9GXbH~NpJSN(n4fAFN80-zecqtGJtfnwbGty`s_51@oDo* z$E*7I>-r4Rfw9zF@DIRZOj}{>9X*8ADXl-l2P73P(HK=$s-ox>f|dUR3>wLr4%3{g zxLtb!mo*szmu_moTXng-As!`<(r{sl6iGLj7Rj}_>L-7ZW#rW*dj9pYok_Uncp#*N zOG%-g!y2)fIma{FqVa?9oZfo+SKRqI$nRAjS7)ldjE_{KwZd>70d z{Zw{x*X~v!`CWV;44zh$&CgE3sD1L zyL`u%Fpn&JFgU_+R0PC-tKo`QN zGB&1qW=E%3b)n=H^jP!tnvdIpVOf8?&=1%z7lz8ksJFWiJozs1}U+{4t63 z0i-W|5sg|R(=Jf6$BtO;l9g!h?Ggo-zWEPXE(#(x_j04p@bekK(Dhq?X0u>1IwiWT zFJ+oEXSjwHFT1*S57cRzS<;rSs$WQw+Ubm1@Nm9`Id4yeUVN`}pl%f05)zpBu{XZI zqA3a_{|H>cK$W(I4dhg-l1{pulJtybK%T>#ee)X^^k%<6VCnBxLql)%Ar=ajH63v} ztfxg0>c`cwSE1^+;&5Qa2rfsp93xrqN%MWS)@hmx=N=JBRxXd@>z4v9^bA_>^YXvQ zMFfBQ(a$m1uAaulgqIlbxOcJ18~Sscjd~!1XY=m65!(WmqNGwsmh^71DESQ|jS@G0 zfzz!5R9V?$qy99=IvaY_pyrPt(+5{?14!$A6TDPf+(9O`9xqT#r*o1Mh#Z?#1)<+{ zhgNr>&4Ry#7hXc!a@!Eze(aw%?OWo%-;1%YN`7}UP1KO!_skqu4&v8bL(MrCpD_3w z$V9L<7K`k@gqep7UHQu0HScAjomA{AYSfKS0_4;;y@Y3_!2J?>cDqdQ5mX7Si|tGZ zK;V95S1d~0@e6#gsf;WPrLJEStGX{435ws5Zdp4QUZRWI6WmN|1cIrNJU873LHX>h zmF=n%Z)6`KqVS%rT2AwaTgYN9wMilK|JCr?^S}0HVYgPzGe#?0x`y-swvuwS3qj+? z1ZSVWf$tFs6}I}!B3O}-=?r21-GLoYB}1C-Q1jTsiH69BGfm9!5NBiO0MGFv0(-O` zZA_GUIiP9SyBBI^$JVJ?D>1!d1{lwra3Z%>&VC#s*cZnllW_<9zNpd%OJb=8nx>31 z;F>g$b&6g^x#^%D)SQ)f8+Fqj%HdWM0%V3>qm-dXhbeT?_(XOYnO{yl?;f@F+bEA( zx~rtU7xl>9^??f_N|3=?6WOW>$ZUSWC8F4#SpOag?Kz{Fd-2kNH@_O%t3NJiF;dQ8 zsge+x-Gp!4S)m5w85}+HQVu!a}xjRpDOt5iRx3 zUaOpsn-`4>bDH!4lRLf`40k>71^3nZOt#otjMmB@AAovqYHxzfcjVNWF;j9Orm5*Q zMecNcD(wX&6Fv5r)S6Pm({#wNf`w;VMDmyV8G1}}WuZ~x03JY@C8$EP;*Si(Tfkq$ zWL?keNt!H}SDmI_fxoUjtR~q+sVBEO;wDY|d&dnl6cgnPRAQ!B_KGFvpb#9P<7aVz zBRfas*yQE>OaE0{>9l>Jza`Rv=K{J5)Zh47!h8a6w2!W@B@%W8z=Y%9{Qc6oxX}-0 z2&m(i3=yr66VeBIG#EGie}}-jzeC+ol8N2g2#`A*DK!d6Ird!S#J3#7S1^~r)KRdE zcY%Ysuc9GPZXV^p!93>wxQ)ZdZBngCTHR4aUr!SF`{H#bpSpyUSPRPQ(OO%Eg}Ow9 zEa{}{%14`TClj5#Db0%5P-QQYgNc=6B*x&fozZvcjwJEV#Nn$nLQDvj*;XdTgaMBk zmp#RbJ;4Ys6vV*gD1<-)>9q6DkUrgwdI>0fx5Ag7tK zgJAvEt4lF&a*DDb0yx_!xWsCL+fSNb5DfK0@?(*mi&l}26&yOOj|0R0Big{UjJ6J=$2^! zd`Ivjew!W~YmNCp{4KnPkR{Z3P}Ble(x~+~;y4zMA1OzQFO~DU_>W`%reuOZj*u~* zMTpHYdn79ei}-I4m~dKr84BNC1r8vU5XG;NgeQ6PL+Qtx%K`5f9)vhGM05UG;`f#Y z<0^q+V?AI-*3-#3X%mjWlmu>^+U4%67;Q=$NkXcf>n^a{Dby(?%?q*^3WS4Pw{~VC)iqDuAgMsMK|WIN>YnJjIM>iBR#q|VZe-- zRJ_;}X5?>^=+VFW4aE?2?7;{VASqOec11~wo6o(t#ed=e7e@1J3DC?dDfI96`u}|! z|F9mA=vtg=d|1;tXyP*Q;85>ld*qrd*T_ndn^v&@#37}7F9OotsYbE>SQ%5@qJhk< zU?<|3+@Hrb=>xLHA$eNHFSxFtbhTR^Cv|Ztk2Zk1kZfQR<5@b&zSU8bX!Ny7IAvq-QBEI%d$Dvx4}Ll)BWa&8E!-QZ z5Ddc|bkc=3#xraYmH43a)3?Gxj$1|0#^|9=;v8U+nx@B#?Orv!l&S?Eckgeub8%Ag ztC`%~Bb>x6Sk{UKkW}=G2Fzd#6X;>l8LvJAyFV2!5^5akv&vZ{$fpvov*ppPQhg0-DTb7lXsY-BD{GGiiJU`Je%TQaT()6)y(G5-A zyJKNTy8cqj_*MGa*Hs3+mI~>eO%nf%KLb}zsRm+ZRkAb{yX@hSsT`)03M%9k!<6~U z##fV*iFFm%jXadI;x|JrBx~1@P$xWe_5J%MD-BWDAx_by@RJX=lhx-?U8~3`gI^K? ztJW_him}o(oF!b&ISe`bs}}2~9U=fgE6CtkY^7Z3XI8la1RFCfPa7=5u~C@$Ut06` znda7h8UB^vm0f~Y1VB(XDh+a2nZ~U*0Hr)9Xc|!@hz+!B4XE5k|6`?~WZ%F*3F_)U z_}^5WNXZ!PVE^|i>+N;O5Dws@_5;29;m)Ccp?0x|q7)0utfWAMp zM?bv+>7h=N;<*DQ686yBKn>CQQQRV7&|Z0GY|flMdhynhJZz^Y3)u=D#4)HQS)DZ` z&_GLAAdJFp{YWx(T1UfI>d=)&*V^;n8c?tM8zUc8vNSS?xkV}uLK6FmpT;_Sa6i;C zn`G2bnIaeKUev-3#y5lAnKTU9z~v42r8Z$fcEr2*LNPwcgLuEm%vN%bN0Av<^Yq_C z4sj7*Luad5&vEYOvLnyWmA=;+TO-`k_^j0P&wKEal1#G*9~3F!#t-)^@A1F4b1L6N zJddYCN-|8qh)y$IRk^_Cd*WKjI^DzEA9$<{!=BQfln;mtNk{M3kbY+vP^zh#2EoO1 zB5RFWGk#FHP(j4NFB1#lJ;HV?cIg3EB9^sQV@qbhJASu2Cux*l@n7{wkJlEyG24xp zZB67@k8A7QL&Vh6c&jLAD+PL!QZG!?n!Q4!Y5;vz5NR}V^<1s}E zOYbq8w=u#K9Z#7<$*0tnN%wYhVomRLm0weUOL6T4ul=JCJ2RnZ z_*n8jQZz*&k2>Fws7g;^Gwe>vOh~Cwuw|55zq_f_vsNh7FJZcmQ~eZn^#hm369hNR z&6_NZ@1r5X$rF#DY7is|%raGkQaNh$XLAFa(L+&$@f{A(qKF%2wx|rBTRjg1W4k2@quA>pm~dtbe659y2&J8RkDbw5 ze10&wF6tl+WWGO#Qx@bRkcUH(2T6XON6)i zDUpZDpVuoa%x+1Xtl#+o3wWh!)qBdZIss5O$eq}+y`J`iqXmonnkD7wj@NzTLO#eI zIJ}|MBuI@hvj)|50S4<9sx8T@%S(kVk?8dmr}*SS@kxK7p0&v7D4hESg@@keWC|bt z!O)86QyyJSMg>G4+bZlPm=yqhz4ao3ZRbGGv8peDDCuC@lzlQV5S#y-o-DLUn(yTr z&F#}`r-}Lo%-?C}2Lq66muu0v1(3`7fJn^a^rwZlOESe%lftVPTfF=DI)TrbNB5&h z9k5K#_C||wEYMKhjEE)wUQld44|$#XRJ|ff-DhYl){@x&2^=0U*rQ-ESA9yzqDPKH z>`tauZ5q9`vSB`Jr+WYRblK^L*%@5}AElfNtzMol;1K;13Ku>Yb>DI(c5Avov_IFO z4E*9P!*yyIKd@bk3Lb#1V%c0{YU$OrBX-jGcVYXkMma)=aA~q>^?reVFM=qWi5;%E z0;NZsq_tEL+i@ zTN7^@r93hv`jFH;xP#=qG32#)z|pK$nZEIPdhtQx)5f^1PisOP&}PKn|1Y!pS1^!y zD`X3V6C8auS+`HOe&#N=X;$O^16wI-Nre`U2?#ks27NRy+j#)8G(paD*kQ#qM6F6+ z7;Gg)Mc5o+7+Jk;aISk>`pB(SCnqmLuK(#{_y9%#eM8fYl`_^Y-vni8ENYy=*)BLv zml+9gWONQUF;DqGcmrDg{f_j18NEo%Xvb;^F#V)C{sztyoXsxX43~hrWnb2+*fp_z zKZgJGbt1nLn}9dk2_UGS*3`vlI560Qfido+;whhP=eC78i&R4L&fBljKq8jr+UjYm z)lg&>{5dfq;c1dh`U5lTQg8*Cb_pk$N}L>kJgx!n zt0oDzIGx!+)elfY%WI!EQbsf3DKk+Whg{CbJYzJsK;o%a5H#WBz1b>-BC`S)v8 zTmJRyR9&)JWtHOA3Ay5%HRqQIFRH;&E)>Cah26b&B2T$t24$1r82XNAt?dI;eCTjJ z6g#|b=h zfO(lMxO<sh9_9M*gS)2=TBE>8p^3}aOjA8Cxt(ToVxn@g!__B1Ba1OFVP^>rU^cEu zH9QAfAUH>1LCxb<_vgFxekDyL`nHMi>19^%BFL1FMUlCW;iR=D1Zk0SRh@l?uN-Zs zpvLmQ^L?sJm1ZAJX0AHsqVP_X)^3u$5P_lq`L@Zow1JLkCaL^0 zQERp*Z3(7TK~Xq&@5ea;T|R<%5rAmjnu=pBjc2fFGX}?Uc@{V|Ny7wOGB&dukg7-_06Uh0Yo@aZM5$h3=n#T2t9QL-P*8d0DNl zky>;O+1kc@_Jf#i<|&~^m>#Czg9;~j`F-(zGVyrr*{{341*jc+%!F5H54`3VE~nIf7U~}8oV_#gl|!Cfqe5^a7|U^UYy0PKvO@1 z#Phy{vB8NHFW2q(w>~2uMXKR{fk1LMRO9Y7XG9k97EmvS2n#qBAgtvE>lX(^$AooG zwC{fz=*;%SXj71XaTAdFN1G;sF(DVh~v!&J6 z@6k@1S=yzs?Rt=j*S;jYhZ`ZJ!=LO9Tk=-wN3XKcz3yycDB8+ZQ7+&Os})HxR(~!f zz~zM-jXo``ag$B(z0Si)wyIZap+m*Y^xM0~-lfAoklBKRC|z?D>^mWk81(`ic93b~ z`5YkcrEx(I`!m}+FBr%;G}icLSFdrI2Cy*e-(bFcm5}1KWgIKORCHubCJ=6=i4H>On09*x)zT}0Z21mCzGFKEaR-0ve3n+vih z?uE?VdE1Z(KN>wwh=-W6Tr25y-v%GIR`Oa+7mbxcF^Xz>XpqbXqriVyi_0o{{#$)! zIMd~3iEV#44Gjw)hM%WiQ2^G-Iit*~4kHpgVc{Wks0tb>4*mPiuj_ZpPvj=OTEc(K z3v6xdttAVRq*#be{Mx5Xh)y{?nU|1HM?#le(JC1#u5|#a;!0oTB03bIPawpFre8a| z&)?}|XCZH_6xRY8Or^h9W!qF@EFTLg)+hs>uP9(awNBq>(2)HT&OpT289&6P6^sk1-?3_|UNy(Qk+3 zfcQ+{tFqP^w|%<0^rqe3q`a0Y74a=}{67c25-Q?Wy5Ewif+4>jkrOM&-<_m2K7)s# z+ss&nG<&p!F3_I}LP&jIMbAKJGFS34#J2#O0FSaur{T5V*qa)J$78wy|qu5V6 zx7HpUuJ|;3KO${GQVSA!ja%>SZ$DzHFPI6&qm|bpS)UejVw%2kcG9$jH`^*EFw)Y> zmMb^Q#3Y3x$Et6))>aU++~3kyNh41i<_KfkB5QN#<|hP}+`&_gLw~=`;BH8cioq8o zFE&xrBZeQlYf5wl1aRrDXOQ2X&5EqGplf*NN`)6e{lIr2nIDo=J1R6R0-&I&*c98; z*lC2EAcayK^+;d2L?Nw_l~`k;#v!s0Qn_m)`cKwJDNLAysgm=z`o!=d6RYglxYcfv z;D5nAWAoyame8M$;c+X%P^0XfFjVoTAuv46adQk7{Paj#Q1UdFHS!;tis5osm==G79_GTYd0bN6LbiN&oWJ6IU9qiX|u!>0m2zB=3x=wF5lo z-VglV9L^)w)W5nA+;ZJ-%eua(@J)q8#<0g>oC9VKQ+47Y9X3l2JgJF<(B*u{fc#>- z@r?v_gv#6iHDX}i!WC^ozpSK|+)?}T+GI*;SpTQn4DoT9ilf-BygpUKm0vE0RtL4V z0E-|VDIO3zK%4v*DtwMftx7(zx(g;1s=#&tN?d&%j><&b)yR$qcdtp+c$p2gy{v2R zSu6AAcZb)IWD9pjr+r2s@>?Vkabp)C+O++wkCp{j=FjyAWV!{`c6J@bw~M*|dTQXsodQ<&nePL4vQ%Um}>cVsJrBNhGaowvjVAHi#42E z)%`GTr-#S%ihpW^FT{lqRy0n&?Y0~V%NQqUw9(JDd-ZZ&br}|7emHEXwp_%e)C&mjOCba+tUYvfjO z^N&rughbP(vSoIxD5y~)0hv~Wj^-Ph()$OA)1qB;pP6IgVOov;zq7^_7twdvG^mmX zp?YJa&LAYC6d_KV2weEDz7LO)j-QEOJ(PcsKJmav}0&>Yf;KRMtClxBw zYf4x(<`Ergt~5ge2*op_xz&)0DK3DO%iRtdKq9-jY5v)wn}Z59jJ%nXvVY1BGl`b9 z-rQjJCP7kSkL@*F3K6_Q|KY>jg#@LRC9}tD9u5l(1m^w;53%FABk(_0MbST$wGJt| z=bXXpOlPq!e6vhw$X9ktn#z@!guUhdVq!z;Y%rd2S=KBVHWE+2SgQ|AJa(-;IIcv5 zhWD;@NVm`!WC;v5>6x5)>a$5Ze+a;u6`d;v&pL`{S6I+gNR7yMZ|igJ`LsFtNhWdJgg-OB!A&4Z zAe=A$MGfK6uj5|;u%i&#_S;Gv(x-i@e!>V^A{em_RHwDu@i@$$N_M5~AVqb}};HNE~Hr=H?0{++3@SH4o>vqXZm zq^O#p-wzzm0!8dUYBf&6h76{3OPd5xPX=SxtkN6F(ADNQ|F}*s|0{KutknN}`+8y| zeW@rM-0UI2*EHZw!uY-#C7S>+7JBC)7({)Wx&?i_I*NA`c;Sfv4Q+xKSIb)iF!H># zvY+5hT{!N&-D`O|)iHYKyG;YAKO|wwapxf{oy>Qj;L7}R;QV#UEj6z)lTM&AFxyz| zp8I&Q8Gb{sbNaccG(Wv>y3mQbk3V{n+FA?b)}f}sk-@gus%RdpqIEKU+D0?)_vPkL z&nu57cQA=dQ5S`7!B`D&DDfW1iZGdN3@QTuCKeW6)RBdVnmM zVfo(ovmrv_OK3J5Z&A11WUCI~m;RnRR|K)u+^cu1)hNJv<9;&4wBkfWO7-~{;Bc?K zGbwY^Z8XUJna!|TwJn&`6wtD#QMR2)pk~mW?`Wv-Np&+eAi?Y3g6|PNxE5M>L65tu z@j-h>%Q8-urpHNP&Nyx*>`H5GNiehE#z6Tk1FT{q;ZI5V=Xz**c5Kue`=kbRA_<=K z3Acms&+N|H$>vPKHLt*#@*`&tZh$%#MT)@hAePxEH8e)H7DLs@SGxtAv%t?Sz+!*7 zC8H^Gf<>PZ-tMcTwgs>G#GreQD7piY4vf>cv)$90DEH<2cq8a9@HvJE)gj8yAPVvv zb#{j?XsWdi1i6VA)XNm9tBvC-^B~#{Kh-kYDm0)Iw3`9_p`j!tJ4;^`pV$&CXq#@= z9VR{B`oiDL)Vd8rr3WK54b_*-R2-vHETSNv3>ZV7$8fq_7#wK9(^XJxC2C3K_el5a zq1V^tH0Tf=o_kMfX5)g+q3r_(*8nfNc>9z)8^y5}{nm1{Bz*-^{OV!gGnj%hG-Qf> zi<%fi?g18urJP6~&w$-nwC|zp?7%Oebwd)p%j6q=!>#c{iGqxloTf5&qD|v)4k=2{dNko8cSnnFhX*kicvFc+yn$wsZCWk+T^FiqJ8Ov#5HNT+wMxxMAWW3@ z9?42awt3cg>P)!3@4f{3m_W9|JFk>Y_%S04#1*FONEK6|Yk`FVLVx>130xPl?$eOf z>~oQ%V47J|Nr#rAU?t^CRp)N%X1kr}@kUp^axc!)daqCiK2Y*Ah2|`j z=)GWp_&8OrkEyAD4x}bd8A81?qu;W~Td<(9D~OpQ4;ONFOshpzbO)(yQ~Z$C+1jQl zb;|_0{9|cggINZkzu|0zsb}XWUk-HjyW^I0v)g^PwMk&_>b#^-LOsF%?MaQXiUMVfXh6DA%?BqMR=c%4q!)k;`;=>I$ zdEkcnDl_QLrazw7t>ZB}JhR%1F~2smv{`{i2HEOSV1GyV5pQr+b0WW&G-8?EZ%ILU z8)66RIM3)!P?C4?i=YoOY?oI~*p?r4hazBLiMoR%z^` z*{GUmlQhde>gOw!`8I7kOrj+BGa z8~z&P47jI}xHR8%>7%)2uywd?w(6NG6z4XvOiPK+s3^ca(z_a@GJEzKKoi3lQ?`Z{ zvy2$)Z&{lW;3eA2Qfdg)K!fUG_-|4F?-qC;L0s6|5oISF_ra!wf31J^Yd8!#nVMHX z*2M~f4-%x%+%G;zG`kdKr(L=F4GYeRDv^HC7?OS_ZEhO4meG&?nTa#cL#4p*23I?( z#vM_7V0YtTv*4tHtlB^ijQTH8%_W*MJI7peTfSnW9JYIf2<;}C^(8*JR{wt?%alA4 z_yqQ1i)B3rl_@h0NarK8+h1%&A2pCTfoEa(xIiT@7gJOsJ^Zou8-hnV{9Cs{`6mADwBY;iD69qL)zhvnn z*dk)Y(9DA}AMdD~ISZ3Kvm@lGwuwsArKqyWOlFVSB0qtldD%Np z;wS`g$`?izkR%30qa8-68S;QG?;_LBs}R6;@|)wjy1hBlrqtyZKGXQoxr)B4J= zCw;Kh%S^~mV9?r)LzhLZoClCjZOe#e0x922Y+=?OVqwECGJYdl_p+WWY5FX9K68|W zLWkQpedolY6#vmA?@04)$uotztb0(9Ex7w@F&u`2O8W6Z5TCcupJ#AdJz>}d<$-(! z!_?A=;dh8xXO;E=D_m6D;Gl54js?I&*qTGXMW8gX>Kn#dnAjg4Zy(`{lL|;jPkp#4;^NlNmX!?Nt6FvcVD&7OB)No)dDDEvY-hqd3K9M)d}fd*R1Ee@5Wg%=zPEXSZTFNp zz<$Mh;$W7KdwIF< zIAlaTAeVj=_^bHimT{=LVFeF8`s5biv+q+PeMo#=0x9p>av!cxU_S3Al@-_1qHe zcVi~_;eOb`V{9e0o4y(a16@>D1vt#JSNVlT+07gC~|G*%a}TVgR(zg546MeFfmwy2?rMn_Jd z35lF9=zT{71Z3@~FpWdGhk^>aD;0iT=O|Jn#+Fd8QIb}{Tk2x!jpAs4KRWB_wh=D- z3hcdjEJbPyxz{RMTd5_xj$~#WRAL*uSzHqpKYe?^`6m!h53Ep47;980#kl?bONdcV z)<5}&0#Y%mtl{^c~$5(R5uQB60PiLj!TRq7dnVW~A-$alh)vgZoW*9C+EI;Y( zlAtlyb(e>&(A53Oq){;oEvW|l`LA6GaZR#isC%aZq^Ho4u4Y!DvmW>cDQ@9Ci{oIk z<&(B2GrSJ1r`Z%d2L%q+EfBw{#lgeMG+Z;!R5C%D5J?jdkXZu3u z^x-SY+!Lk1POrI4YE+zFqtDP@`{$;Nn7DNA&OURvN?8C2oB|~ ze$WPJ37x8Xjjqz;TSTUcWlTdPi>#{RtoqPXcTIlO9e)5$Y#7Rx!tM&DtIS8L!K3Y3 zyzzG2Nlm5+Cdn<@(gg_Zu&>dBq@+R+b>f)|nC%O^SGXj7|6l(? zfNm&{F6ixS)xV}lmCXgrf0BAEKEB4|w};%~v~ zo=*T*Da57=xf>@=FP;O7+FqtCZFdq*HUY{-t96bZixI!Gn~Xd#mPO64uWO!$r+n_T zSHqe!7;7$!H-t(RNp@B_$0Z;O_3`iQ_Cf3GjWTS4xL~LP{KwU%!3^I z-r>e7PBjR7pot9?9pq@#+#R%7+ZBrkU%s$8WpYt`wkk}RSe&184k4@yty9OP5aw{T z;2KU0H!6MrZl)hLOr@SX7K`K9VDUD^eCEQItiXMF=l(CI%q!Mibd|8|IT}KJ>e;wy zJ#vooPFR?+Y8WJYri**Cb~u~0be}qETVk;Ln=XnH38d6|?EF0n3LbuI__NMG8XS59 zOSYFB$5<_;VVvUxFADP^44ik)ms%ppe9po^My53`-c&FXG!Z!%=Ler!!4@FS?$?)V zEC9hQ>fo;FaW_4Y-)O$Y z;Q+|}7z1u0BLa%wOev^c541_p%0Eb(-%Kzuii64*>znsJ}@OH?Y4J z7>A}lRH|U*1HS?7r^RM-cWl+KE{HiK?tT%TxYiX`3R_5ovzTf6$<48Ltq^{6(XGRd z%-R?wE?-(bEj^4mKJ=SX2$A=n4S5(~%F zO{_lt&bhY)ty#XOxaOwHOG4wwT3IwbL%UFG3t))ue_1-_c}xBZ3bHO|Y%to-tpL<% zFserR!M!w059puaS@u{wEDvCt`#|wfy~h5~?gIL2S%Ov@hJXzI277_M6iD(&>qpu> zx$p-R=ma~1uAM%v!1ZWMh6^n&$+Q{3f1py?#bx=P6rK|j($)M!^J8=RV%JdI)c+nW zX;H!KpNXz*@p#rrZzgpvSy> zY39|eqo5bNqwkEJw~F}>sDWg+J3e4HypHXiQ)5*3`)dUBg%7!t_>hy0BQfA#(^Ed4 z@3iYFFOr?}wzjm5L!i|2Rppv*XIQkYb^MNGMzLME^tDysPoJJzNFG`4OX&xhZX zP9of4n`%tRwYIc9ua(npE{BA2s$&anY!wjeDeiI=+99xiWjA;RWBU9Zddq#IYwl^a z@EVk#wP2B9EBrU3r&iAhgl|4|&=0EbFoqR`3=5X%Q`PwX(?c4HAEKlp>&EBt`h6d{8y@gTY;3aU&;M)J?o z-5kq8UO(1U5Q6fe{d1Q8r^zGN-cz6p1Ir-9X)IXZP-A8MY{bcW!{ILKj&e_flRJ4I zh=g18^dmLcR~4{oVT`>ZTYs--G@1DLkt<97z^3g~-2De_VQOTh6Z4)LIdsvI=f!7* z>x^Gx$SQ)q$)LCo*Y-Qw&LY^LWi zV;>~bli`bM3`V&zF|l-$ROc>Gf(V~Jzsj+4sQqS@V>E+Fym9NyDl(n6 z(nVRw^sOtvHLU+a#Im>%qf#5vmtSEAHa{-Tqf|PqOv`exl(Bz^^iQbvI}im@G15P@ zqNU&Nzf$^m&2M*OyIsW(>X4F4`qW(ngqV8=CzbAU7{ z+Dj6%ynmhO!3f@C3al4oBJvTn) z?kRmKeBaA#Sdk4=tF*|fd+qviy;_stji=lKaxW)Da!UD$sHYd#(2{XtHjefhKAAf^ z^+STxqK7-O0tCoDw8Mv0dEXGTew4Ik18VqI%LT?$ylYFTP?SyAzW}8;woG?{7s2%f+U7_7;{}I9x9SnOxhwazTZ3pQlQs`A~SDn8&V%JnbK0!`RK&*r`g>+LcuauKt zu5Mxg1+RJb3UM_ntm)Fz7*P5fSBfCIL8TGO3F5v{ssU&bCIR9h~Tdv@J)i%aO zAGO^jeuPj9Wm9u!EYHwT4{3L8jui*+9atW$3K7V51{XH;+vmDz*}u1XLVX&l|C9NV z%L%=?*=ED!8D*(LQo><=M3j`evL!xrq@SfGor;QY5Bpruz{1uU87OT-&z%eg-5hiizgmj=~G5;W$ z000q|L7F5Ogn#>&WO80jnC^)f8Z9eXR%t%w-7LP)H6!Pg{#HjX>8i{vCaaw_)Y*Ci zm0C~l+MUNN7RnD!(zs7OQ%fE>iZi^=-K2YLJqquKmnc+<-bJ2)H&lYvaoRDMZXGrSMC*BkXMmBy^lo`_e) z48*7`F6@1vUlI=}Spb66=?}wSjS--QtyL*?pod{$3YUsxz^qFq3?za!Wl1+wsjyT; zwsz3piKDt>44R!2Rn=3bA8vVor|y^7!3eeNEf0# z(|WpGqJ-(fU6vKPDFLjf`y&oJ1luip9w1dS^T1O|bt3QR?IJ ztM+ItHYWalMny;iFh=0^fjJMH0cQvRMPgvm9GFNi=?VVkVM6ZxBw&Q22+>-x_oKaL ztIXgJkDC(UfO}qu8p&mCd{ZuCcAj6E8ECu`T5uaZ1@0#62 z!2e4>OONE=&3*+i6}vHqq7Pcsi?Wz*miN(pB-8O}M@z^kWG;rCX;1r1#P|-JRYdD#iphqi*v1^%Wczr65}{fN>HCR zDx0}SS@JD7GNwm0NLZzja+3(Nghfq1WomI`L1HCXR42c*k~WKwli^MpRii}s^3afc z<~(*KvHSm*=4|${Vi$hkBj0KT54opBv9N@~?bK52I9+VqL6@_=O^EHOBlTgbme0N( z*XR)E=^yJxpGd;LZ0LGW2S!8 z`E`Pb)$6xN)!bu!qyjCvRNgIe%yS}5O}C7QP_I{GL@*xN8Fn#KM$b4}W!YB_xVj+~ z4h)7Ol8;O@v;)A68mRwU?}~RGC}QaDj|e(kTVP?}B|&IOtZLyiIDbl-6D)hau2tM? zKrRW#O48vv89b)a;Ji-JD zyc{GSV4lf9^8<5*jlOC8NVM$B+BB@aPHy7jHLY#zSpuDJ!koRwU9*tl3{t2`E|3*1 z;W>~VbU8BjYRMt?UW0>5s{gzVu#~_LkxWT6&1#Zi`{-RK`qec6Gmb}isu2MK9<&PS zahnU4ZqjK$Oy$>ZLuND2#(yT0grs}c^Dh(V)vvs%`%|7^4q#Y~Dfa7zsEVn`PN}RS zFV!Gel=1e+6RIG7prtnT$xn%6v^4<&#US41b(q;+9YL-`m-cX}V~3n)mcFXHEtVDAB#dp4*JwV=NFaV~4W8S-PpH@~gJL@6>qMchATCJfWzzU8kyw?Vb) zx`}q2C|mcoD9*e=DHbOn{NW^}gCZJGIQGt7%H1p_1bu4@1w?FA?ppm{5VOB-2w?jO z17{aqlgZr0ND(x)XU`(j=Kon`7HW7jr0kKNJZNjy*smyM@njEcWz@v$v)l?HeZ8}x zf2sN4eTswqQXamihDtIg>oW&LeR$+2(=rV4n8+w_4G-PI)fvUC?e(T*BQBVZNAVx0 zJ%I*+&Wogme37A2x1>-M%b&=CKk&EGvk;y4U2o*UBR?aOVvNx&xv9`7L0`sqllHVm z^OJTGfji&{} zxYLniNpc9qjWE#lyOzPU_aG)$7a2(YVk&+urxOaxxFyJFz$(Dd+g;&vFnJU0C)Z=W zsM8mTaDyND0S9_-R>`mGVWBE70Yw->tDJ(1%Va_&twc=+zYNC1z3x!vGBx3e8_qs7 z@V0bynx=LZSOObyiqPO@mmLDw044BL2~s_Vf4}KvOa<$1EKk3M|4AwK85^EvUVSI1 zw6ZjEQaeB~(;^LAjIS}xqcTU>iDJbVGSo!o0$oBbn}g-^1RB~cau^5(A;T|r-9jBM z7=9<9M}el35)_g6iGehO9R$xO#As-BpIc?=V(bPn6g|at-@8;l$g@uWGTB1?&OLw5 zD3DWSp3G0)Ph(Icy5w~{o$m&I2|P+){u-N_l9#*=-91^u3N0$I%X{E6m_W=Vks4^C zl|bO-p#AcxN>0e8gdBR2JI`zjWLYgb{xF%#W$h?{nh*qnC51K+<;HK`FPocTyKEQX zVB=4`idMC;p)ktj^BjE%@IWMer7n?MLy_MT z%(Ts|^K}pZ6u{y%Ki^}2hCRnVkxHQKTA9}ST6#Rk@bRp!t@aet>%Yc77Z*+2lpRVe z-~L-8P<5BMo(T$XFLD>@y4r`3=7~^KJXAQ|g`RVlTBojEV{ex`z}8Z*rwbP`)(cjp zsrE?W48m3>3q7FdL>S`Zjzv!cxKz;Ed`I}PniZvfi<^^ch_}+)16{3kNzFI>vL&uK z^NLOa3LED?M>(UM3lDB1!z;mRz?pEz{acf&xqtES4vFJlON8DJq=6w1det{cpQNwg zPaeW+9I_4_=E_=m2I8-dMKiJGxKU%h&xu`6HCJF!wnGW%@7eibNn)Tapi; z8T=ucYKQY9GuYY?lh67$1CSJ?*2TBs(V`QalHrojBI!h=X{ZF3gHb6IU-k9^G9s7jE)!@r@9J-3&tZIVfAwyZ?~ zU~bOAsY!DJFXk3jXlS8Z+FP^=P~O;y1PNX~nPCzCnCHCmum<)+Ro3C;w!ieX8OkPj zC?((=#a11H*DK#!#}ZqH;eU@-6=rvv3-C<-EvUz~zuM>{1Z?<`o-qOzwidg5M0JQy^yH zl-R@QwnB1)2u5PWL&6m3G2!PE77N~*E1fKRW0M$;agWw6S!ED5g5nAT9*>q8zI}nuFKx^K^jQR{9OF? zy{8CT2$DLqxQ&CTuiO`)GZM7T@E_#y6dDQDtrcQ?S8}??PJ3zv&UYCmYHOKnkIGxTK>0h3sUmUlv~Ed;O+6BAu!8>zPWGM~^PE8{7I=s#J6LU9UU`OVw5!gP~yTckAaEh0_0bLXzZ z)Z2V4&=DSeoTr2i<~hNQmChkJ>m1ISy5AZ1;@haoVDQq9@WFB%1pR0N;0s~5NL~LN zAz3O+Y(+Bnb^E^JbtlSuEtOib?2m{Dnu&(@I{cmO%y0_+&&;wGkIAMj0!B{16e4$ohIo9=x}{t zjni*pJ74YaRU+QbzQbkPYX%?J)9C+=bv1I$vH~4in$R1)wk&!TV|LNmBy{S&%!dVY z)3NEi-C9D}2=v_Jk*h+vK*R(G-~g3(bA)yu_Xf^OM-FI3uS`g+F}e4JgDfl8`u@q<-)d1)+D{7!4B&t;UgcP z=ZetJqq5uc>=JSf&}P&PE`ztWhrg2%uYc;J*EBY&(%|UdD1d6g?8>I%+yRX;6soA1 z$JLP(v*lfsUSRNacCH22&~YXlDGW3qGMrpJgvx0DYEUKVi=PnG>m|-|EC5h(U;Q+% zWl5Yz*}Hpg-Dr|B9BasF=w7h(RrRiE&jBUj(pd4q_~33dn@Z^gs0F7j#E$3ZqldR( z%sK(E1fxW8^sW0SwKa6M7T%_A|Ckjvi?ht`kM8Hj=&?yDwU4Pb^4D{?aFqCILCQ0AfJHqw?_Ow`GfU(=R9 z=D{_JCjiL3jRmIx2UbYEC}VAuu!T%?KcnA$`dpFp9bl(2<7uW`&$OwOUR-Y*6Q@%! zY}BZ15Z^2`P7E_21BvTNHJ;sgWdRHC?4SI9RtP_Hw_z!LG;2g^b_b;GpRY(lE4PT& zO7WG@<4qCDAjk-#Pd$Za(JVo+^9WxI&0$#=-X zOfG~+)iJ*KPKqnkNuxr$@Qd&KP<}*P5s4jJSw`-su_nXDA@P@CoS>cgWy0zq%e+>c z3et9NUkf${2Vx1K{hsWjcniSR+%r?hrLx{S4ak4b5-_$$g2g5obWW>Fm>OJ(U*WI_ft2c+B=X|d~KRWD>_KpEm zO~#lpJhOUOBsQ#ukalfrao^#U7fy~V5(%sw!@?nhD&%iX$geFA(!CVaVMLEqpiupq!)QL zbM$PgBTgeFRY{vwY->S!UP_lnoF;WIfo~yG)0Q)22sRFX=YwOk{46)>rD?m~aBO&S zjlz3NWC^>KZ?v3E-~Zo}%H~#_3efVv__VFC?t*T181)6Z6;W7z5Zf6pLK)In?_OG4 zpew<%?7*`fs||nEJ;r9kE9KNZ*??Ap`&JMhSeNJmcy0$zBO6bDJSoNKBgtKO&*1~X z)SHoupnZ7V^v63h(unp=a%BlAQNvSaaB`QDy7Dc_`eF<_b+(3k2yMD9j_bKz@_dim+EJKYzzO=47dgW~ z_#2WF%zT()zE>^A7q1}W?9gugCd=kS5@C}P`<52Nt{5ysInJKpM8qW4R4?8Q4NW?6 zHp(P*M?xzQK=4-e78+mJTXT%qA~!6SvqMAwXdfSL6>tH~LKeRl6yMZCmupFfRLHQH zrv-)G11y7J-tt&R6R^bmBg*t9IBLnZ6gSe$iOAqf79hn=feSGB5zSQ9=3e-7l+$?k z(3e}fbtbm;Z>-1ivEnWKac6ny-ssH;6WOgFIvJb}d!GsVH#F3Awr`u(a^@#D+6p50 zE7T7n&btuZ5*VZ_*O&rgxlF8OGI%xjCzF9%$Pb>GF~u}eolNh~ctZGz0T1T#+I>4Bo^2gj zTU*2Q_3_cHisYmFl4-oD+Efm8kp?V5fY%-+f+Q&~{$$>)N{&QlUJ>%|E{kSgw@rpF zQi=#_ks@VRo|8J6MysN&O~ap08gUcLiTmY@Br!e4A-GdU6sNKnxr~(JY~|vEEG$-? zav6F4OZQ5~ICxEvem5BW4&P{R;B|*@(haI$x)?`VK7&yBr@`v2oGz=|gs{>sJXPwF z)@0g%d2v#Hnn6<;{wg3t-O|~SyHI}dDe@p|2&}tq$cy0qz8tH`oq@RwpsE4pRV5c% zr8;>_u(2ZD(ivhzz_mnZ8;}>`a+#J&9wFuOkxINu*TXh=0#;HWM0-vNP*bO@FemfV z1MoC`<}1%TH7lVW`Z zN36_U-1qr5TfO_#OHKJN)l%%n(OZ4UG74q|HhIp{KPHeD1!66od386CNoo>AZ;<6N zY@Fty<7LcPUlS_W&PU+dvFN1hrT4>5{vHQt+{^ziz zH?_gaoN%2Ua{H!KS@eVkH1v^2Nh?I0+h{vqneep@hegs)ra(xm-&mRul&*$@fH&}Mq*n?fv93$&~_Fe?+>v|rSP#BhU zB&c2<QL&ggf>r#x(#7l z_wRo4FCnBj`ggS7EmaD;$ZOf;WVRI{7v?XdfsL`dG?Rk5M%W71wKvd=R%92V&a=Iw zLh7Z67@5~{6<1$ONQ31E5zrPwa}#qjHr4#7!q~Jo@_WWyXXg?Y7RsOFNC`+i0`3Mf zkUU=bLT^XVvwEt5w$y%_uydYA?vX>WBg<^Eu~!_ZUwIV7>p)}d5s|jMIt!dQ)R_o; zyJ7h(IV@%UekP7HwBrr{SyMYA*Dj5g!+3#b3u+2BNJf+kx58C%nm-a4ens1XPb|bK zw5eL~N;}lYy(Vs-FMDz>+?xIqdG-wpbc*JtYdr$j4#@{!tfL#sE12`&vMgrWvXFUl z35fY%r@aY(ta5fN@!#cXUy+CHVCv=S?Z6iZQC3!v`=Vr78=1mT+{{zc__CM6GTYvD zE{wO>d@mnTaoJ$k8jxU`I+5pcJ^{|!bkq@RWL&xBuSR2C@eOcxf3Q#AaV}9TL4T`x zAe{(vo<~ILBj9ThI0<7!>Bi`b*S&{E4|tiE+sh7nI^@S=Nse=Q9KT_#$&(%%c0-4# z(}5>1@P(&fft`#~bLZzfVi_Gwbo%4pwE>0H_6y`!3Y4_F+V%V65f%TYnC*)us8W__ zkB@!=!K%*5oG%btBz^;UbgyG^7}O)3uO4kHi|gQ=f?>^TRGmbv@@J9L(D;{NGb4r~ zKTAJF+Cn-UA+9)v()7G8EQVC%rk+UWN5OBnzJ`cq0V0xc`sHYsFkiM5t-~-9L>o`a z$YJu02^`XnEDLyQ?g5uSmP`0T>XTzuK|YsC(YPm~I^(1SCO9UUW7j&B=r)1zI~fDRMOP+8kgu9RUshs%GzzaaJBFZFc`S=Ztc0W{EHvcdpc1`T z0ePi~X9m4vXL}qx^k8I2G0)IkyTL6{V5y#tgkL9;E`wr$(C z?LKYWwr$%sPxG{G+qP|M`oH(idoyn)R%K;mL}X^=hgG#|W$fKD;9LKqxYi#UYL_eP zBh`lS&-hy3@${|qGWa-BIjC@(st;)1bPR-sl_Wr)Gb;y8%)kpyuT!Y{j`I_k?=pRW zm{;tiN@o(MNeu`O4=eJdVc(QkIaKzT4KE^k@aCBdTy}}p$l)BN@J2+<{qNQ+cOF28 zoZ=?ScN%qfWWiLl@JsvO2f=QacqrJ11L%MUhxhj2 zik;G~ZJgj__HwzjWQxe`tR)}SRJ_Bo zpk4p~V0ge>BN01-kM4?S&|v+V0wM*r#&QOQ0F=0MfAnKnX=f`fk143gbQKR(`3`OX z`oeUZrQdD2P29v+Jk6%zBg9bOs{|AJiu3M9abF?(~31GimEG( zT0J=UIX}cym-~lW`*hpU;S5GlDaQAwfH*!60Q4mm8oh#2$2?QXH=e>eIh_fd? z@J#v>c1FJlMU~*Z;dxun#Dc=3Do;*XXInc!Sw=U1?$r0e-k8&}l7V{X&*Uu61@1eL zkjeM&r9mMLkq<5SI;_-Z;@-|zb!X4yz)qCn}3$Y`;U)f0P&_K`qu#tGnK zm%?xCw%wt9FI%%O1CKja(%=-V+(hyZfmvU@+z5r|mm-KIz28~PD}a#G*ovNAbFhf= zZ9uNkS)7AN$QkX~AD2^#guY0a3#2F0ZbOIPAbt+3#inXy-cj34YW0@a2gTWza2r_jZAq17d$*fFG(^niBFbL$$I)7 zWj8Z~@BZ28@Z(pqJRkp12!j!2P-!m+7JgL+kRa!uglaz!In;0uKqGfAop z36w0TY3e}-Vl%}+;&g;2tHdRUrA6Zd2vZM;VTyW_^V5?WL;ujG4w30UQ0};egXo$UFHJ+sB4?iZnk}(*dU)Xo7-}) zreHd6C8rhtQ`t5u`8(|(+z&|agtUW5U!U=~ka~+*w2A~K?iT?=U(=PRd6G|Y&gyu0 z&WjqSFaL#uWpvmyP*g~13X9yeRmcZOsqQPSdY z1PfZ!H4KU4aCxEZIS2g@eW9VsrjB3Hp3tlR3-}^elS!VWVYvr?%?uwfQvPluZt`xi zY|Prp+wJM&EO;@sRqX=}RYV8ZA~F9(e||Ix1Galm=!*w@fIc1x-*cSKFb@CU5pX|h zJ=k;vCYtJCw49I5fRL)#ye)AJc9dFjvqVwHG$vE?jmp_XL!PYktvb^@FsbxarQ zZpd+Jh4*5e2z7Uv3O=hk<)jk7suqPkqVDL>`5>{VKdry$@MolYO<59N~X)qi~UM&_mu_Y zJkp`rQ`;AfA%C`}x?1XNF6#$M+H`_(Zf(U;8ns*k!JlROOl$MXj+lDKYzC|@Gf+QF zd}QsK=E5X8LB;Nx8;R@UEp`8%j4ujFUq=HPK}l0@TH!PRSg!2wtrnJ(@QHkOgb3qt(K=D34aKoNXS z*Jwom^_;f_KDnf1n8$Ts;YHPZ;>S@G1o_7S1FXzvIds=g2Gqg-jO>6A{8VI#`sW&b zjN0!HCM`)_)-8idgpV0^y`} zVr0r*J};%?ogRK+u%K1s;`mhaAv+1gSU%L*BGajrwCtT1M`E~pT_{fZ8+k;<{iL4O zq|y`6o2~w#c*4>Fk_p62xRdMzc%IrTQw^@?NEp$J7@+Oky4 z6yd11f*+Us`yPUpt*6nsVDwF_{?F7Sb1cwPTe7oUq7pbqX!`EZ~D2#pEb+KW#x1R!v@}Y`Np(t zWA&}3MR6Ig4PzQvb|wgIAKGv3L(l-HUr<{XQNCD~^Q#(u#mhX)XP%uafA z_veQpB>r=Cec5(e0zPaZBD)2s`rFJwx~r!SUTNX603#FAb%)+dL={)}q_U@^SFPGk zdFZ$(<=?f|$=k{W{s1Ah`{37b!pE_&Z>yr2G19s%c8DNr9RA7 zLOA7|uxMVtTd=e|E3_pOAx{K(L^%^3uxS>EDorIZ#aTuhCnh<6I&gAeQtn$Y^44qH zJD*nvH@F%5R6BN+)q3uAMAlqH>XK%^4+^S=FC6hR>h9_$&rfC@euvYP@= zIcRboWvxyjKI8siy^ZZkIax1u>Wwu15SsFcABxk?Fldz1?ojb_aH$g$Q7$f}>5Zxd zM5o16&n)b{P^cHSViZVY*s5ny*szC<_y%)3GO#+*_odN&t}#&-;W7#2O+d(2NnE)U zmhI0y0aI2O#3J?w5jH{V+UW&exUGkoAP@6E8Xz8P?V-Rt^AP&T6}O6My(Bam-#FAGO0^>OD6 zo~irF>8?trSjLHH;~NfbdGB7;35iR{Yy9BHM$`Zet*8Ed0q4Ps5df5x1w0Mthq^1w zDCRQ%hZT#bJH$lqN2jfv*UMc)VDU)RG!;9|E$Jfn#n|5o;MPpHpI%zkm`$P_4quw{ zLKx;_m9{~~-drIP?iUv9I>vA zvjt}T<^@Ym*K?riPVgSprdK#ldHN1wC)=pDO%O@3-cmnW2(WG%l&B`39Ctdm(K3QI z`7=8QUE6PHM_nGtYPL!5D>(j~RFq2z#`ejIm2D+e+I58+=GNOVKXx#+))#>tC&gD@ zmq$HrjzW!rH`vpVdHlV|E$1hJ0(6@^m{Q5Ta^+_9mc;9yv3f@5q)X7?eV^#*cD%9` zo3F81-B3(|M6)8%D1ve|l{1;qrpM$sop;l<3P2PwcRIi2a<1tc_^u0Z)-Xa<;6pZq z9HESQ{Kb0;s`Kc>DVH)}Pc^#=&;t$!-w3)zeNiV9bb_tfR!1)nX<|BFxP+;_gKu(_ z&2f}g`02_WwM+fKnrmd+-FwN}P{k^9ln-Z09*+QJmipizGv(B`YnKW(n? z3|+KAb=}RCU<05?^};W0c*mHgkOfFsZd6`7h(pF#;xVyd~@%bB#jHzmiLv%`n>kf?N^s& zfvl8Da?|p$6TTXXC#$P+Q?Q$tE*YHm_#f7B-n8G9>(Y3iU+Q4X{Hh0ds~JV)ObUk# zEyQpz0&AW;{0Jow!*dT7wVZ}h30ozwdX7dA#6}Z2EQee? zt@O4&ya{E8_I;1CU!4SEm{bc4c7%7Il)K!rd3}F&1qSXSAll^LRo7CJ(vpntci>8n z`0nccLKVKZ;)CPa;yzW8o7;7VN2)t2>S@w1vY0`JE{bQ?K}m@|<*DD5^+Ol=7Gxs# ze$ia;X<*zr40B;S{toC=v<*g$)>B4(<9fc{ZgwolE;u94EO-a1JYO-8zn@l0GR7-~ z;QiqryIiPg#vu24$R)4|lGyWq8@A;bC#p2FX3LlLjEi3H0?%?P-L(T>pR=X4c)f}m z{7VBIrn0OkUyEo+Sfb3DM?E3W)CrQli+A4fpoBWaQxZ)Z`q=<8dXfnR_h4D+N@(~Z z7mw+!UNWnodH4(A#vA#Z8rjxNMO`5noDig2WlRq z^0BR3-(s!_MWtW;Ua@*{^6>;PF{v^hf+XO)xIO{NKFJjq;33+USU|@+T~~)LC+~jT z0~B$tLv~yKBzAI@2_G#E$^#}+V8?TW9HwYVJvD$7*M9=ITNspD+zfRvFg6U{wT&7M zL88;FzC*>n&~x5*BXl&=h2oM3OQf7E{gEOa|4q;Pf{Nc9U8O9Z*AD_gy=z^ zlUBYW`ps_`Oy6VTLw90Y1&NfsXrX~!pHJ>r+Yuvh*7#%pSyRVnr# zH7DyVEW)D^0SF#49 zmqVJvev^T7Ql-%=?kP#il;|F=3U1)m$XgtQdVlDp$Lekv2R@Vm_3n;?PtlZ$JhTW} z@q_`^gYzebx)X`sG#uZD<~6h5z^Dr&`1DTn;9e;dyAS6a3yeJM3bT#4a7cY00`SM0 z^WSYl^-`c#%2GGOrupCBl7G5oPqF!djcgNBR`aF=XFAPMG(d+KFS;ZK(YcGyt1UNo zQG0uCAL_2&xZe+--jC1EPfZT9)eLP~3;$3vP@|*zPshpZS0^kW0J6mX=Ij}*7}$tE zCB~Z#xIu!voDsy=8Dqe87z&KHNPA_lL&{Hcm7jcj)pKyN=ybQLJD3|qA0Qp9Ttrfz zZj>;%Z4!PL4+MlOU0-D;S8#o1pa( zZeF;-WK!lZyc0d4Z7w3L8rKpo`h6F1SLO_pA<}@4aa$wISrPY*#;(_r_0`` z1qvmZC`!V;n+ZVdM}qzNr#5BuzA!``u9XpLJMFe2Re3=kbLhLyzMUHtkvJ8vn50>D+deT>q4+&cPu=-2>io|n!~(f8t|vG@zJ51 zlQ1Tuw@GjWNWaAeCeThj$!Jg=3lpYUa>HC{Q!Ujq*KACQF85Gd*!xZ3m8Qobv{1wc zF-;=E6Y(fF@F-1jV{IeG?+p}oL>Q>6vJ_#{uouz!8Le_e>d zcOy0vpGW{7BtHq* zl>po!?IN#H02M{@TJk`T-v^aki-VtoA|75B1HWg!-{F6Hp{i3l=Wgz_Vhb!hpie@{ zPeBwoNEebxlQ#r@eu97zV3?k%(`d*^((NwK4ASCF`G)$QV0sJwTEjX`cCbct@7&exo-E#;?M znMs;ieRUV08YCR1z;kBIKI9_;h6hF4V4UI(Tpez~_fQ1MZ~an|4^&TvR-q438+(hrgMHgA#_ z&ryE%c=ILB01~(e)Oi<0t+TR^my4c3-`_+j#_Ae!Y^vX0ZUwUZEABxDQYZlg9U3 ztqY4TdElsJx*#=ZGyyi+I}t_CL=zh^Z|B45oQEQ8=g;%tUo4>CF9xID@cLwP$PmzI zbgD+*a|+QtC@X>mCxgjQSvvbSj5pEq`Wt6Y#==nbF(E#);sSEO65YJp6Q9KO<- zzZ;oxK{B}#o`7I`jW!R8KbKxo-{tpKSm3n3nvU1m>O&E!Wot2u6uUSuTA!PxYTycm zwBP_kLUR`uNwOq8Ui{6sn0F4zQp4i=~HGNE5lCq`miYe4Bz8X!ty4>xUIY z)Ely<`1e817KXea(WN9 zDEsG7G<%@4Q+wp?PVgD1g!R%Q>PewELPPYYGVT*@q=dLH0Pm(k$8=S0ST=1$0qBo4 z5n=e~`lhg^xv5V92d*nov!-_=rJc2T4qlqz>_&{oUPz`>!4NbQGwr0sF9G#+giCP& zCh{x5CY6ip$#{7%n#4f&dPYO$E0oUOus+vZ=jT5s2HmKMaAET?PfbbNd9(w$p_}gX zOEf+3^G8BC5f?;8y)as6nC`Y~*i#c*WI=knJX~>CSZqfftFzpvAM>~GyN+wA2A`21 zGst(?qx?XzE4RX-Vz>T%PHJ4V#T%_(5v$fG<4c-^3 zuui$fUvz=D2p4?AVZ;W-ZHS9)D=S678{pGrp<~Nm@q78DSX@B!Jk$_zFxk^a=Q>Bc zau~t3I_}zQ)1PJG`{oPSRuw5Tf1eziCre203P&2h6!BxLd7PGG7y#NkH(D+BK7o#a za15vN6ZNXO*!!L=i>Z$LulYnYeGxV*>kPNhuOlgK(^fHjX8VvgwuRRS3gC>2MZM3M zalwHwrmD_V2Y0<2e$H{>eK*g4Qo9Ify+buEY?rQ9%D@oVhrDv1qoARB;yolKDP@TF zCd%ma}k&P_4pqV%w(BU6y1H&Ub^^Z5I^%1~AQ| z0Px>#{H@qa;c{qIDrL$c362Y2Xwz?isYYW2H~=DJFZZ+PxMc9l_OKxny{{w6*yQDH zuE=}`UG(Nx#S$*Qv{9IYJy4G5O8IcW z!ygr1>qH@ICF~jsjCm1hjM=jK#If6`(U{ms{VWQ?Uy&WTvJ5c<5t_u->Q_GjT=s2+H-=zaRB4IxQf~>w=*Wp!V ztj?GK*!xNF%4vu0J(uR>E}VoM5RnjoW1h-%mtPqqDC3n&TD{X76M>d=J`ldl9!cXY zj)7^iM{1xlu$nJQH^cNq37;CJc-GG8sqPDS+ojx6M%XYRCc+hRXtRssh}f)on_YEl z);+nEbd`Ked>*Z6)#MGpF)W{0CC<9hqvtcm<30gI#D{bDGx36}IB7m1idSJFQr#0n z(0qQ9RwRZ9kh%&CD8YHLO?NM$b3SS_A#_%Y&tb&3;kJ%k*Zk9rdIqlZxd8Q-E2vY7 zij_;~xK;OY`dVZFC3fZpON)MuC&osnbk?@SPopS6t-HKMyIigwwNfoK z_y`Y@ep>3H^tDMht^i;~>oWHDLEp`88U)C|LS5DZSGkF_aBB1|@ISoyTKJe&>DPwD z4Y!@}dX?Z}$qFCDP&7>d5m&RzxCct5iDjbM+`D5io~u%vBqB54_ej&zIluS8(`#@o z@>(HQRgK2dUgGZ-9CJ=Qbk-q5ar&^t%_8|_YX@zhiwDGU&-?<*TxXWy{}`)PiwTzLSb=g-s%jREt$rx=d-C9x`JJIKN+ixecsk zar3-(j*rP%HHL+aA%|`d6g}!WF;Yk&*B$HJei8ppni&ZJ#$&oqkl@VBqrnLtX2$*I zc7KG^7_B$TwLa2g3Ksy2uEs|m^zi%A^&Wh`(qK-);uHw!ObPbli}EhvXKBuga8yT< zCsJriV+dZIcDlD*DS43jC$& zIgSgdau6sR@Y!~Yg+)igeq%61FzR17-PRD;>LL%ofv@d@U3Ed(BpnBLPd!#bQWs($ zKn{=a80k^@RGx+S$u~m>HbeN(Z3KGbiAEML>@+NGD?8q}0Q(l%=Zu^G=|q~y=YJ>M ztt@(X)VLcy&!LWnetJKV_xj0VYgv7f!|*rehl8yei1iNNxsX26H`oG04>E&2xu!l( z(+BDIT`oA?Ei21|UZCgHXflbH6br0HDwk3=_z*0d$!U(M|1;xV>v6zUyF)WvzS zaX~~5cdd2e+YwILLSRl;ei~k06(Gc|>wS+V)9LvfX8cx`a|+KO(#n-4?R5CS*3?-R zi5<08vWmO=^ml>|mDPe&OQOea55Sey2+H9z;Z>|E$PY_?phR~^AS*G;Iw6@wmh=hM@gE&6`pw%&O#8aj_Uir#2 zt^h`_8!7XP9}__Ry{u~MU5zr_@|@VRgL__kma2OB)BWIZ($f~^m45!RPf0f7ynR6x z=DO+Hpu{*`Pp;6%RQKayJcK0o01!eDUrOV7d$DsrR~11QN$^oL(A}T>g2e-rsuxpO z)|gFC5`Vzny>h+rw)aTc-sy&E3R;8@T-`v$ue!vo=lF6ATYYe2zZC?)f-YlhWMZG; z8?l`sz8_yY6Gu1eob7;(aB2*!A|clks_j;fc|YqhY+i{tVv1<9mIJy7bP{X|3nnh` z^hq3%Kb4V#zIl{Y+BIaUW@)-9L)l0hgSmH0f`Gf8@^Y&UVu0q&PEg%r)~xA-y&R=| zoKBTEniEg}{!ml$WH0%jNz_u>{S5P>z8VK`B!uvopIFIGfkIju`oufx%C~VJRQEk z&!DRk#O>{}X|kMHL7jJMc4k&Rf=yu2Y&UW&yE2cP{TXPU>BPsX3l6}dW9&F*d)te{ z!t=7%zS7{5AE^iFIU}zKNUr%eQPJ`il|K%#0Cq zSrg0Tjq91b@V;UV>S|u;;B|yOLjCtxJ-SQMwTJh@gJyFCgBF?wqv)mFU_6+vxCw6A ziLY|Y!-HLhjSe+9TLvtYDch?DQXab`RJV$z;Z0fCtKY)SDV$>r4qLnR!kFu*E$qmW zyLt-R>3eF(4?~S9Y(8BP!@PeNQ)KNj32sJ<@-~wty0smjl_wXyxJimpxFdlj@+6?! zRSV)sJ7BRO>j=4gFix714E22@Mzp9JV;s6}$QimO`KPKA^M!HyYf=t@qsvk~W6gu3 z_`pKOI38k86}z5ZJj@YaT=<`C^Zo(vl~S)__Me(%N@O=TVnLQ6IJHAqH8LM!MaP_w zvpL2`!M6kcT`Oj&29UMhpAp`sr40Vf^{0` zA4`zfXRzW`=EeEfRxI2W3BQ;|yMPy9g0hyR<5$-W0rS99^Rlo^hVA!JP>{9o{aqSZ zKzm!fV8GxafRW-}(!>YA(gUnrLO&I&aDs0|#FX*1iCs_o;KV1u+2D$hD)YdH9Q(xB z`)i#rbtJY=vGQ3BI*`$z6=2m!QzDiWQqBZLV`>E1W0c`UANMFm1hiCv?Dg0Y zJ4;Y`(4df0gzla+Q1q@>pHUgnJEHnuieS=Hq~ zwz=HMHEQLu6SNXE<0BxAiQ9*03BA(CJjASY5js)s4pcd~O$+SdF>huk0^92i+xGTNYFltu~q!s?p`oGLSZtef=)oM9P`@8 zq*4Hxt++6T@6q%r0S#RbeHg}zS+;`sv1jEeOk>%<#lc%D{S!hNuK^wA)&V2Og6H2mb~il7GF}Z!1Zf9NyJd-?1EKzT+H@*D$}B>aq9WuPBq~W8?BB&I z;Ak@hF9r#y_iEkI=-BBhx2jfBZRC2SAy2p0B$xG9gk0jSz!`vRnwU2yNe_d~6s+Px zYmo9v5jg9CWYAu4#MJS&cwOqG($o&>cK@tcBKCGue|Hlujruei0RKV{1cbn)S`kj* z{W|N&-Om?t%YCXnje)|js1u~gt+_hKQk|;xGG#alI=L_CDnQE$#~b%(((8L<3*2>Y zk}5NJAV55`B*3l$RNL*i7|#(;Mbf3S;zPt z{98_~G}vsOCd_%>oW~RpCWdQDSQ4Yk4h)!E9*n0WyI403K!36V4O#^ZvVb=w1G>RYFK+LR)az z5cz_*_b8^mj_n$U4?JD(FoTr_tMpEfIDkPAAR*GcjKA^~Y$f}eH8XE-{rtnu$Ao=Y z0?g7+`eqw-;0sU}v0lG!>>&?SLL8d>?=`g}j@qZX5Ob2VE_E|IFTT&W7rhW4vF|-n z;yF^|2<42(mxgj}ZfP8ct?6Stb~Dh?6-kr%1LlawpQ+`iu$OKF3}4}9_}CqyP29+& z&g)OVohrpa3~ZKn*rmrS5no5~w5m>IWizHu0rVP^@#cKhT}bpL`>HvK+AbBA&I;0ApaAD2(W z=DY1Z<7V0H`AFeLBD59C$foEd=J4n!;RbRaOU3r@<4>#_$ARTTeT8?FxBA1j)8YQb zvUY%f$dD}o7xkJv8*%mD4^DuB+JFCD%I7JBRnQD?~GiK?;GvQjYh{e`oh}gjZ+os5QYoWV5cyu z=|`RpbD$avpN&*SMC)BwJpWO_Q*4j~jW7$Pm1Eh2ByeR7Z{;DyNtI>9=~(9vqbaer zCjPO133iQUG@tr9j<)2ytqvzzx4S>Tdyz@g39dJB%518RcS6>-4ZiFe^WI+KW7vGE zWExvg9I~40>m#MwF)YNF#2!-D_{wjNM)|IqfLAC+*<+D6;|EGa{MVY%ATC&~Bx=Fx zl78^y0D6Mqlhr7*)B|_~cv8Dr%Ob)Hx?rFC?oMfa0eYBoWlGb>$O+7HC$hBCwS$Nv zZnSuV!TLn)2eYs>kK~QYw=F8PQcDodDvg)ToMDYWbg_2}DJlFjqx$;5(oP6vkZ0hI z)TeC#)XRy-l%hoF124a>j!L@@U5C`RUoFY#I06fz3YeBiz)K*wV)Gwrg8=}fxG_S! zeE^^;=AltDW2k*T?poQJGG1Jn z85N_GC+|zbj`Qktu!bJ>$)<7Ek%iFd(W3frr1~2v+@z=VgF&tE1pY27*<(hkCmvCw zWPKA@R{^eZ$b6de2X_F@Sc((F9l^1RT=~xtN$W;5=%yaAvmrJ3w8y5?x$Q52eul^}YB~91UfR)spg#=G)Nu~6a?0<3tZhyX4@5CuHqP|nmGC@@x7m?B`vJ#2RWHC0s-WK7)rY^WJv4myNr~kc4 zm?P)oSk+kYuN?8|9cvx^6pxz2x080uNCS-+ID(!1Puj(v#llRhl*GJ@P7_aXGHw!v z3ej#d)AG=lXd--$zKBk>rW*35UsVck;3%mhJQGU1?8<#L`Xp&|)WgKolvvLK;S~wu zCd$Jg%5Tbk@gL@j<}pi$J|pZnN+*;C;WHF<)cCi;6zJwZoBKX?^ee?am2S=$^i9HT zPWUV=M_SrfF(R_4{jCanZ!L%iH;0pvJ}~5Iyk=v0QDa-SOnN7@jY~$pR&YthgpJUL z4WEWxC#Q##>FeJH6;74yD}JQ7%KY0!_`SWy0x1?S71&a9$kr2$*nEYqI``3ey(-f>a&IyNsWI5W-pEfu#2Qv zo{+`Is(Bu`UCs|$!@aaL7`J5-&LYz|Mum;&C5_nHkyyB-0O{8Y@kg1?lBxmLzwgMa z2C=02;==rNpKPon>hnE>mz}X{w#LMHr6N3QQPsUE!k4>w6^vHx&d?Iwh<}_k*Dhua zzZ$Ih9u95almnVJadxc&2#vOdV4c2_j0I4BZHc)Ky)=6@`_Rq+ap(j84ci!~V^tg-Ls!|KiqSliN?klZBKDF8;fT zK};#Go2S$BGfoz-AC7HL`muS#dyKCJ#+iIohSjhQ4EVX*esxWf&SJ-ve<4j$V%OLJ%4h^%l?Ra+!;|QL+R^Dkt&gG3=-$%L) zAG%PnYF(DuWAjfdIL9y&ZJ@%4`P!M@q_a@QMNBWF@7^LVSm6L&qFruk9*6btX3z(tYbrfL;e1WTVuV$rGQNbzE zRqZY(Z@_V5isx;Z2womF5U&b9S_q&eOv3eT`ruuAB)R-N7pv8ynq5LRco<}n6XxhE z8BG2AQkYCs*>DaVUilArKb3U9ZKx|1pewFbOKA16qcvjg4b{v|SX1%&r8Z?8R)x@x zaJY@T35)N@R_aJ1&#O387^*n0zw_bi;7I zedg2-1&Cd89?Caf!KsUI^T+1>^k2HFnXG;*_{KcLN*{=%5l#IMyimdz4dl}#2n_H`${LzIOpCj74l+poAhN?h3g}W!(4m{)^mahWbs0Ym31Bts?`s0XpOl^} zV7~7>L@;ifezXkww4T|Z>1ayqVI}mB+O``?R`fs;^8|YXiwgLZRi8Xys?m!GCkegH z=*Z}RHTVJ%eqDFO^t1kZ?B&4GspD+~JnUuf=sQ?0g)R2R&bVg-+?oue6ZaJ9`>x8~ zF`zIu-9No6fd z1Y^$q4NsjRRrclqYlfN{NM*0NjTom)ejP|eJzl2IxHM0gRplE$YB&HRD)Ew<8w%>; z9F1M$q#?T=xHo2DR}G>Jt>-pH@IXnDGPFBhN0OwFMH<+sRq&raw(a+`RWL&Yxlvz% zYg-{rXfLtJ4TTL2o?a2D((9SmXe-fm-PI7?`3BUUj)QE@Rf+I9^`D=*!dF779T)g{ z-kokj{eoJL?jEm7<*E+`0e{pt;J5wr5H2acs}3JC2m*>dt1h?vQPc0&88eqpLta;g z%uD5O5H6%TP%KEtK%?JX%_WP+FIhoj{i;;f7)#T2T~@|}LT3*%XINuFHrsB6FBjAG zz@d19~PZK!=vYnQNIrYM6?$wPP#ZhP(nq(@j z@Nkm!KDY1&8=X~qnOg`|nvCTj9XlRh=b02|NCk;i;1{W)sP;N(rm8QY+O zIoOx&rqhlx7Oagc>EmRv=LFPPouYF}d4E;L6AG$;A-YgUtSI{cmizlGSS93H|6%F| zlkH#~caLbb-Ob4+o-_mLT}Trl@wi_2*3QMODx?5SMh;-&zXR=bidz=?WZ;&p?u#CV z{Q$)UWxCi+_xN638(!y|H~NwJ&Jcg&NdNA_pt;H-=)m=Ot==pxdTT5)a=3=-l%IV$=Wk*Ke$OlBc(=eA%t<(-8cET4irc1O%(MWhp z_{|tU0n9BE@v}$lw~0HI`cA0lp1l+C%gk{%RSwB2 zf7+dk;QcMx+e9ZOyT8j!g6Mv3w|3Q7fFms)_w3_KK?-|mP2F6^8OZr!P6j;_h;zO~ z!_qpnaAs=n>l51mLHS62OXGh1+7i8CbC-khtgL5Cem%hJ82y747+Fw`PZf4I(!ibz z$=PPQT~MU=TIJa`3c!?FU?M71k|rddypwW&V~Toqex@HheuTDV5|l9Qwxjhqh2fVm zL!G`szUg2}j;F2iqw-_=yFUg1=g;b@nX&O0pQtBL9So{YExmJ;hM}suaBL zOfz93Hs>y26%gJt?-6i9q2s|9op57RJd^eW4Rs>nzxSP11@{9UP$58-D`5-O3HHft z`-Y)3$hfB4NfqJx zS*A)aFeA9aUY7&SswxW6r`iBfNDy?0E;@gGJQHF|hdVB?uOo&Ne4Iu7wp*>wa zK5x>J;c_fp*B18upNby2(dT~T#Dc&eg~5|lReFz}kq7Y@vp8c9~f&nZL;diYln3#kivGdhtV0+WpYP+jBs9O1KDrDcGv?^KRmW zInQc&E2?t+QJr@HQrHUl;A>o=Hk<*SGpc*>=`p^~m^NslIU^Of1QJ_`N7+LNK__m!C@bz@ zpU6O?6vtn09$%hEdtl|KdlH+fp9!86W(h`HC=@2;&PZH%8pJKJ5{Yu~L+P z!|xW7VxF?|Ja9?XFC(;NgNIO3_XZ)^NcdZm&0C>G*gF@K3;Fjy?%b9RE4E7nhdRfz zA0EsdO$ER>kyR}|37f#8Of^&8dn>WOSm!spT@HApK zLw4!A&^0#D=uB9OPca!aRSW-8f&#l_oj}|Uzno-qTeL)2oX=6!G%+VY7wX)IbAyde z-N14(X&h(!&VZRs?LDCABV)sdq5pYrQF{cU~$00R8y_}}vXYW$l2 zkG8=7j{N`AfB*o1?wuVCtbU!^&Q||5C*c2{^M8%@_5bPof6Vj0BL62);I{_-EkLvX zWgsyzc6R#Ja7?V7od3)6-&guOF3A5}Gmxc?v4z2}kHE(G|6DuIuNj#`W$?dxGMXD( zJO0lY0In9sCjYnopIX??(S$(O$M52Q zmoD+^`A_+e{C^_;+x-7HWBgzH{~yS+0GjH|n8G0c{~?1+I*?v-(;sDk*}5 zzUGfU1`uXoP$(_|#|)6k089-G91semp4kG(k4Q$ya{~E5abS#rX)Y)YKg dxCb4`=L!T$10#&}HBcSUo@7ue00EFaIsl^$*pvVO literal 0 HcmV?d00001 diff --git a/apperception/apperception_benchmark.py b/apperception/apperception_benchmark.py index 2a6c0bf4..fa4e416b 100644 --- a/apperception/apperception_benchmark.py +++ b/apperception/apperception_benchmark.py @@ -10,7 +10,7 @@ # Let's define some attribute for constructing the world first name = "traffic_scene" # world name units = "metrics" # world units -video_file = "./amber_videos/traffic-scene-shorter.mp4" # example video file +video_file = "./amber_videos/traffic-scene-mini.mp4" # example video file lens_attrs = {"fov": 120, "cam_origin": (0, 0, 0), "skew_factor": 0} point_attrs = {"p_id": "p1", "cam_id": "cam1", "x": 0, "y": 0, "z": 0, "time": None, "type": "pos"} camera_attrs = {"ratio": 0.5} @@ -43,7 +43,7 @@ point_attrs["time"], point_attrs["type"], ) -location = point.Point(pt_id, cam_id, x, y, z, time, pt_type) +location = point.Point(pt_id, cam_id, (x, y, z), time, pt_type) ratio = camera_attrs["ratio"] @@ -80,7 +80,12 @@ .get_trajectory(distinct=True) .execute() ) + print(trajectory) + + # plot the trajectory of each cars on the first video frame + traffic_world.overlay_trajectory(cam_id, trajectory) + # Get the videos of these items # entire_video = traffic_world.predicate(lambda obj: obj.object_id in id_array, {"id_array":id_array}).get_video() # entire_video.execute() diff --git a/apperception/lens.py b/apperception/lens.py index ea68cf61..59df6775 100644 --- a/apperception/lens.py +++ b/apperception/lens.py @@ -1,3 +1,5 @@ +from __future__ import annotations + from math import radians import numpy as np @@ -36,6 +38,9 @@ def world_to_pixel(self, world_coord, depth): """ return None + def __eq__(self, other): + return self.__dict__ == other.__dict__ + class VRLens(Lens): def __init__(self, resolution, cam_origin, yaw, roll, pitch): @@ -219,6 +224,18 @@ def __init__(self, resolution, cam_origin, field_of_view, skew_factor): [[self.focal_x, self.alpha, cam_x, 0], [0, self.focal_y, cam_y, 0], [0, 0, 1, 0]] ) + def __eq__(self, other): + return ( + isinstance(other, PinholeLens) + and self.fov == other.fov + and self.focal_x == other.focal_x + and self.focal_y == other.focal_y + and self.cam_origin == other.cam_origin + and self.alpha == other.alpha + and (self.inv_transform == other.inv_transform).all() + and (self.transform == other.transform).all() + ) + def pixel_to_world(self, pixel_coord, depth): """ Translate pixel coordinates to world coordinates. diff --git a/apperception/new_apperception_benchmark.py b/apperception/new_apperception_benchmark.py new file mode 100644 index 00000000..eb8523b0 --- /dev/null +++ b/apperception/new_apperception_benchmark.py @@ -0,0 +1,135 @@ +#!/usr/bin/env python +# coding: utf-8 + +# In[14]: + + +import lens +import point +from new_world import empty_world + +# Let's define some attribute for constructing the world first +name = "traffic_scene" # world name +units = "metrics" # world units +video_file = "./amber_videos/traffic-scene-mini.mp4" # example video file +lens_attrs = {"fov": 120, "cam_origin": (0, 0, 0), "skew_factor": 0} +point_attrs = {"p_id": "p1", "cam_id": "cam1", "x": 0, "y": 0, "z": 0, "time": None, "type": "pos"} +camera_attrs = {"ratio": 0.5} +fps = 30 + +# 1. define a world +traffic_world = empty_world(name="my-world") + +# 2. construct a camera +fov, res, cam_origin, skew_factor = ( + lens_attrs["fov"], + [1280, 720], + lens_attrs["cam_origin"], + lens_attrs["skew_factor"], +) +cam_lens = lens.PinholeLens(res, cam_origin, fov, skew_factor) + +pt_id, cam_id, x, y, z, time, pt_type = ( + point_attrs["p_id"], + point_attrs["cam_id"], + point_attrs["x"], + point_attrs["y"], + point_attrs["z"], + point_attrs["time"], + point_attrs["type"], +) +location = point.Point(pt_id, cam_id, (x, y, z), time, pt_type) + +ratio = camera_attrs["ratio"] + +# ingest the camera into the world +traffic_world = traffic_world.add_camera( + cam_id=cam_id, + location=location, + ratio=ratio, + video_file=video_file, + metadata_identifier=name + "_" + cam_id, + lens=cam_lens, +) + +# Call execute on the world to run the detection algorithm and save the real data to the database +recognized_world = traffic_world.recognize(cam_id) + +volume = traffic_world.select_intersection_of_interest_or_use_default(cam_id=cam_id) + + +# In[15]: + + +cams = traffic_world.get_camera() +lens = traffic_world.get_len() +ids = traffic_world.world_id +print("cameras are", cams) +print("lens are", lens) +print("ids are", ids) + + +# In[16]: + + +filtered_world = ( + recognized_world.filter_traj_type("car").filter_traj_volume(volume).interval(0, fps * 3) +) +filtered_ids = filtered_world.get_traj_key() +print("filtered_ids are", filtered_ids) + + +# In[17]: + + +trajectory = filtered_world.get_traj() +print("trajectories are", trajectory) + + +# In[18]: + + +# draw overlay +# import matplotlib.pyplot as plt +# +# +# traffic_world.overlay_trajectory(cam_id, trajectory) +# plt.show() + + +# In[26]: + + +# render tracking video +filtered_world.get_video([cam_id]) + + +# In[21]: + + +times = filtered_world.get_time() +print("Times are:", times) + + +# In[22]: + + +geos = filtered_world.get_bbox_geo() +print("Bbox geos are:", geos) + + +# In[23]: + + +print("Trajectory distances are:") +print(filtered_world.get_distance(0, 30)) + + +# In[24]: + + +print("Trajectory speeds are:") +print(filtered_world.get_speed(0, 30)) + + +# In[ ]: diff --git a/apperception/new_compute_lib.py b/apperception/new_compute_lib.py new file mode 100644 index 00000000..c6bef7d4 --- /dev/null +++ b/apperception/new_compute_lib.py @@ -0,0 +1,18 @@ +import numpy as np + + +def compute_heading(trajectories): + headings = [] + for traj in trajectories: + traj = traj[0] + heading = [None] + for j in range(1, len(traj)): + prev_pos = traj[j - 1] + current_pos = traj[j] + heading.append(0) + if current_pos[1] != prev_pos[1]: + heading[j] = np.arctan2(current_pos[1] - prev_pos[1], current_pos[0] - prev_pos[0]) + heading[j] *= 180 / np.pi # convert to degrees from radian + heading[j] = (heading[j] + 360) % 360 # converting such that all headings are positive + headings.append(heading) + return headings diff --git a/apperception/new_db.py b/apperception/new_db.py new file mode 100644 index 00000000..4fbffa30 --- /dev/null +++ b/apperception/new_db.py @@ -0,0 +1,381 @@ +import datetime +from typing import Tuple + +import psycopg2 +from bounding_box import BoundingBox +from lens import PinholeLens +from new_util import create_camera, get_video, video_fetch_reformat +from pypika import Column, CustomFunction, Table +# https://github.com/kayak/pypika/issues/553 +# workaround. because the normal Query will fail due to mobility db +from pypika.dialects import Query, SnowflakeQuery +from video_context import Camera +from video_util import add_recognized_objs, get_video_dimension, recognize + +CAMERA_TABLE = "cameras" +TRAJ_TABLE = "item_general_trajectory" +BBOX_TABLE = "general_bbox" + + +class Database: + def __init__(self, reset: bool = True): + # should setup a postgres in docker first + self.con = psycopg2.connect( + dbname="mobilitydb", user="docker", host="localhost", port="25432", password="docker" + ) + self.cur = self.con.cursor() + + if reset: + # create camera table + self._create_camera_table() + + # create bbox table + self._create_general_bbox_table() + + # create traj table + self._create_item_general_trajectory_table() + + # The start time of the database access object + self.start_time = datetime.datetime(2021, 6, 8, 7, 10, 28) + + def _create_camera_table(self): + # drop old + q1 = SnowflakeQuery.drop_table(CAMERA_TABLE).if_exists() + + # create new + q2 = SnowflakeQuery.create_table(CAMERA_TABLE).columns( + Column("cameraId", "TEXT"), + Column("worldId", "TEXT"), + Column("ratio", "real"), + Column("origin", "geometry"), + Column("focalpoints", "geometry"), + Column("fov", "INTEGER"), + Column("skev_factor", "real"), + Column("width", "integer"), + Column("height", "integer"), + ) + + self.cur.execute(q1.get_sql()) + self.cur.execute(q2.get_sql()) + self.con.commit() + + def insert_cam(self, world_id: str, camera_node: Camera): + cam = Table(CAMERA_TABLE) + cam_id = camera_node.cam_id + cam_ratio = camera_node.ratio + lens = camera_node.lens + + if not isinstance(lens, PinholeLens): + raise Exception("Only accept a camera with PinholeLens") + + focal_x = str(lens.focal_x) + focal_y = str(lens.focal_y) + cam_x, cam_y, cam_z = ( + str(lens.cam_origin[0]), + str(lens.cam_origin[1]), + str(lens.cam_origin[2]), + ) + width, height = get_video_dimension(camera_node.video_file) + + q = SnowflakeQuery.into(cam).insert( + cam_id, + world_id, + cam_ratio, + f"POINT Z ({cam_x} {cam_y} {cam_z})", + f"POINT({focal_x} {focal_y})", + lens.fov, + lens.alpha, + width, + height, + ) + # print(q) + self.cur.execute(q.get_sql()) + self.con.commit() + + def retrieve_cam(self, query: Query = None, world_id: str = ""): + """ + Called when executing update commands (add_camera, add_objs ...etc) + """ + + return ( + query + self._select_cam_with_world_id(world_id) + if query + else self._select_cam_with_world_id(world_id) + ) # UNION + + def _select_cam_with_world_id(self, world_id: str): + """ + Select cams with certain world id + """ + cam = Table(CAMERA_TABLE) + q = SnowflakeQuery.from_(cam).select("*").where(cam.worldId == world_id) + return q + + def filter_cam(self, query: Query, condition: str): + """ + Called when executing filter commands (predicate, interval ...etc) + """ + return SnowflakeQuery.from_(query).select("*").where(eval(condition)) + + def get_cam(self, query: Query): + """ + Execute sql command rapidly + """ + + # hack + q = ( + "SELECT cameraId, ratio, ST_X(origin), ST_Y(origin), ST_Z(origin), ST_X(focalpoints), ST_Y(focalpoints), fov, skev_factor" + + f" FROM ({query.get_sql()}) AS final" + ) + + # print(q) + + self.cur.execute(q) + return self.cur.fetchall() + + def get_len(self, query: Query): + """ + Execute sql command rapidly + """ + + # hack + q = ( + "SELECT ratio, ST_X(origin), ST_Y(origin), ST_Z(origin), fov, skev_factor" + + f" FROM ({query.get_sql()}) AS final" + ) + + self.cur.execute(q) + return self.cur.fetchall() + + def _create_general_bbox_table(self): + # already created in create_or_insert_general_trajectory + self.cur.execute("DROP TABLE IF EXISTS General_Bbox;") + self.con.commit() + + def _create_item_general_trajectory_table(self): + # already created in create_or_insert_general_trajectory + self.cur.execute("DROP TABLE IF EXISTS Item_General_Trajectory;") + self.con.commit() + + def insert_bbox_traj(self, world_id: str, camera_node: Camera, recognition_area: BoundingBox): + video_file, algo, lens = camera_node.video_file, "Yolo", camera_node.lens + tracking_results = recognize( + video_file=video_file, recog_algo=algo, recognition_area=recognition_area + ) + add_recognized_objs(self.con, lens, tracking_results, self.start_time, world_id) + + def retrieve_bbox(self, query: Query = None, world_id: str = ""): + bbox = Table(BBOX_TABLE) + q = SnowflakeQuery.from_(bbox).select("*").where(bbox.worldId == world_id) + return query + q if query else q # UNION + + def retrieve_traj(self, query: Query = None, world_id: str = ""): + traj = Table(TRAJ_TABLE) + q = SnowflakeQuery.from_(traj).select("*").where(traj.worldId == world_id) + return query + q if query else q # UNION + + def get_bbox(self, query: Query): + self.cur.execute(query.get_sql()) + return self.cur.fetchall() + + def get_traj(self, query: Query): + # hack + query = ( + "SELECT asMFJSON(trajCentroids)::json->'coordinates'" + + f" FROM ({query.get_sql()}) as final" + ) + + print("get_traj", query) + self.cur.execute(query) + return self.cur.fetchall() + + def get_traj_key(self, query: Query): + q = SnowflakeQuery.from_(query).select("itemid") + print("get_traj_key", q.get_sql()) + self.cur.execute(q.get_sql()) + return self.cur.fetchall() + + def get_bbox_geo(self, query: Query): + Xmin = CustomFunction("Xmin", ["stbox"]) + Ymin = CustomFunction("Ymin", ["stbox"]) + Zmin = CustomFunction("Zmin", ["stbox"]) + Xmax = CustomFunction("Xmax", ["stbox"]) + Ymax = CustomFunction("Ymax", ["stbox"]) + Zmax = CustomFunction("Zmax", ["stbox"]) + + q = SnowflakeQuery.from_(query).select( + Xmin(query.trajBbox), + Ymin(query.trajBbox), + Zmin(query.trajBbox), + Xmax(query.trajBbox), + Ymax(query.trajBbox), + Zmax(query.trajBbox), + ) + self.cur.execute(q.get_sql()) + return self.cur.fetchall() + + def get_time(self, query: Query): + Tmin = CustomFunction("Tmin", ["stbox"]) + q = SnowflakeQuery.from_(query).select(Tmin(query.trajBbox)) + self.cur.execute(q.get_sql()) + return self.cur.fetchall() + + def get_distance(self, query: Query, start: str, end: str): + atPeriodSet = CustomFunction("atPeriodSet", ["centroids", "param"]) + cumulativeLength = CustomFunction("cumulativeLength", ["input"]) + q = SnowflakeQuery.from_(query).select( + cumulativeLength(atPeriodSet(query.trajCentroids, "{[%s, %s)}" % (start, end))) + ) + + self.cur.execute(q.get_sql()) + return self.cur.fetchall() + + def get_speed(self, query, start, end): + atPeriodSet = CustomFunction("atPeriodSet", ["centroids", "param"]) + speed = CustomFunction("speed", ["input"]) + + q = SnowflakeQuery.from_(query).select( + speed(atPeriodSet(query.trajCentroids, "{[%s, %s)}" % (start, end))) + ) + + self.cur.execute(q.get_sql()) + return self.cur.fetchall() + + def filter_traj_type(self, query: Query, object_type: str): + return SnowflakeQuery.from_(query).select("*").where(query.objecttype == object_type) + + def filter_traj_heading(self, query: Query, lessThan=float("inf"), greaterThan=float("-inf")): + return ( + SnowflakeQuery.from_(query) + .select("*") + .where(query.heading <= lessThan) + .where(query.heading >= greaterThan) + ) + + def filter_relative_to_type( + self, + query: Query, + x_range: Tuple[float, float], + y_range: Tuple[float, float], + z_range: Tuple[float, float], + type: str, + ): + # TODO: Make also work with objects of other types + cameras = Table(CAMERA_TABLE) + getX = CustomFunction("getX", ["tgeompoint"]) + getY = CustomFunction("getX", ["tgeompoint"]) + getZ = CustomFunction("getX", ["tgeompoint"]) + + ST_X = CustomFunction("ST_X", ["geometry"]) + ST_Y = CustomFunction("ST_Y", ["geometry"]) + ST_Z = CustomFunction("ST_Z", ["geometry"]) + q = ( + SnowflakeQuery.from_(query) + .join(cameras) + .cross() + .select(query.star) + .distinct() + .where(x_range[0] <= (ST_X(cameras.origin) - getX(query.trajCentroids))) + .where((ST_X(cameras.origin) - getX(query.trajCentroids)) <= x_range[1]) + .where(y_range[0] <= (ST_Y(cameras.origin) - getY(query.trajCentroids))) + .where((ST_Y(cameras.origin) - getY(query.trajCentroids)) <= y_range[1]) + .where(z_range[0] <= (ST_Z(cameras.origin) - getZ(query.trajCentroids))) + .where((ST_Z(cameras.origin) - getZ(query.trajCentroids)) <= z_range[1]) + ) + # print(str(q)) + return q + + def filter_traj_volume(self, query: Query, volume: str): + overlap = CustomFunction("overlap", ["bbox1", "bbox2"]) + return SnowflakeQuery.from_(query).select("*").where(overlap(query.largestBbox, volume)) + + def interval(self, query, start, end): + # https://pypika.readthedocs.io/en/latest/4_extending.html + Tmin = CustomFunction("Tmin", ["stbox"]) + Tmax = CustomFunction("Tmax", ["stbox"]) + return ( + SnowflakeQuery.from_(query) + .select("*") + .where((start <= Tmin(query.trajBbox)) & (Tmax(query.trajBbox) < end)) + ) + + def get_video(self, query, cams, boxed): + bbox = Table(BBOX_TABLE) + Xmin = CustomFunction("Xmin", ["stbox"]) + Ymin = CustomFunction("Ymin", ["stbox"]) + Zmin = CustomFunction("Zmin", ["stbox"]) + Xmax = CustomFunction("Xmax", ["stbox"]) + Ymax = CustomFunction("Ymax", ["stbox"]) + Zmax = CustomFunction("Zmax", ["stbox"]) + Tmin = CustomFunction("Tmin", ["stbox"]) + + query = ( + SnowflakeQuery.from_(query) + .inner_join(bbox) + .using("itemid") + .select( + query.itemid, + Xmin(bbox.trajBbox), + Ymin(bbox.trajBbox), + Zmin(bbox.trajBbox), + Xmax(bbox.trajBbox), + Ymax(bbox.trajBbox), + Zmax(bbox.trajBbox), + Tmin(bbox.trajBbox), + ) + ) + + self.cur.execute(query.get_sql()) + fetched_meta = self.cur.fetchall() + fetched_meta = video_fetch_reformat(fetched_meta) + get_video(fetched_meta, cams, self.start_time, boxed) + + +if __name__ == "__main__": + # Ingest the camera to the world + c1 = create_camera("cam1", 120) + c2 = create_camera("cam2", 150) + + db = Database() + db.insert_bbox_traj( + world_id="myworld", camera_node=c1, recognition_area=BoundingBox(0, 50, 50, 100) + ) + db.insert_bbox_traj( + world_id="myworld2", camera_node=c2, recognition_area=BoundingBox(0, 50, 50, 100) + ) + + q = db.retrieve_traj(world_id="myworld2") + db.cur.execute(q.get_sql()) + res = db.cur.fetchall() + print(res) + + # db.insert_cam("1", 5, "1") + # db.insert_cam("2", 3, "2") + # db.insert_cam("wid-1", c1) + # + # q = db.retrieve_cam(world_id="wid-1") + # res = db.execute_get_query(q) + # print(res) + # # w1 = world() + # q = "" + # camera_node = Camera(cam_id, point, ratio, video_file, metadata_id, lens) + # # w2 = w1.add_camera({cam_id: "1", size: 5, world_id: "1"}) + # q = db.concat_with(query = q, world_id = "1") + # print(q) + + # # w3 = w2.predicate(cam.size < 4) + # q = db.nest_from(query = q, condition = "query.size < 4") + # print(q) + + # # w4 = w3.add_camera({cam_id: "2", size: 3, world_id: "2"}) + # q = db.concat_with(query = q, world_id = "2") + # print(q) + + # # w5 = w4.predicate(cam.size < 6) + # q = db.nest_from(query = q, condition = "query.size < 6") + # print(q) + + # # w6 = w5.get_camera() + # res = db.execute_get_query(query = q) + + # print(list(res)) diff --git a/apperception/new_readme.md b/apperception/new_readme.md new file mode 100644 index 00000000..d1b4fc26 --- /dev/null +++ b/apperception/new_readme.md @@ -0,0 +1,24 @@ +# New query engine + +## Prerequisite +- Create a docker container for mobilitydb + +``` +docker volume create mobilitydb_data +docker run --name "mobilitydb" -d -p 25432:5432 -v mobilitydb_data:/var/lib/postgresql mobilitydb/mobilitydb +``` + +## Design + +### Data Persistency +### Query Optimization + +- Each world will have access to the list of operations done since the first world. +- Each item in the DB will have a row indicating the world id that creates it. +- When call operations on a world, we do not execute them right away. Instead, we collect these operations to be a sequence of operations. +- When we call get-data method (such as selectKey / get_trajectories) we construct an SQL - query based on the sequence of operations. +- When see predicates, we nest the query of the old world. +- When see update, we concat the query of the old world. +- We can add cache on each node to accelerate. + + diff --git a/apperception/new_util.py b/apperception/new_util.py new file mode 100644 index 00000000..f40c5ccd --- /dev/null +++ b/apperception/new_util.py @@ -0,0 +1,194 @@ +import ast +import os + +import lens +import point +import uncompyle6 +from video_context import Camera +from video_util import (convert_datetime_to_frame_num, get_video_box, + get_video_roi) +from world_executor import (create_transform_matrix, + reformat_fetched_world_coords, world_to_pixel) + + +def create_camera(cam_id, fov): + # Let's define some attribute for constructing the world first + name = "traffic_scene" # world name + video_file = "./amber_videos/traffic-scene-mini.mp4" # example video file + lens_attrs = {"fov": fov, "cam_origin": (0, 0, 0), "skew_factor": 0} + point_attrs = { + "p_id": "p1", + "cam_id": cam_id, + "x": 0, + "y": 0, + "z": 0, + "time": None, + "type": "pos", + } + camera_attrs = {"ratio": 0.5} + fps = 30 + + fov, res, cam_origin, skew_factor = ( + lens_attrs["fov"], + [1280, 720], + lens_attrs["cam_origin"], + lens_attrs["skew_factor"], + ) + + cam_lens = lens.PinholeLens(res, cam_origin, fov, skew_factor) + + pt_id, cam_id, x, y, z, time, pt_type = ( + point_attrs["p_id"], + point_attrs["cam_id"], + point_attrs["x"], + point_attrs["y"], + point_attrs["z"], + point_attrs["time"], + point_attrs["type"], + ) + location = point.Point(pt_id, cam_id, (x, y, z), time, pt_type) + + ratio = camera_attrs["ratio"] + + # Ingest the camera to the world + return Camera( + cam_id=cam_id, + point=location, + ratio=ratio, + video_file=video_file, + metadata_id=name + "_" + cam_id, + lens=cam_lens, + ) + + +def video_fetch_reformat(fetched_meta): + result = {} + for meta in fetched_meta: + item_id, coordinates, timestamp = meta[0], meta[1:-1], meta[-1] + if item_id in result: + result[item_id][0].append(coordinates) + result[item_id][1].append(timestamp) + else: + result[item_id] = [[coordinates], [timestamp]] + + return result + + +def get_video(metadata_results, cams, start_time, boxed): + # The cam nodes are raw data from the database + # TODO: I forget why we used the data from the db instead of directly fetch + # from the world + + video_files = [] + for cam in cams: + cam_id, ratio, cam_x, cam_y, cam_z, focal_x, focal_y, fov, skew_factor = ( + cam.cam_id, + cam.ratio, + cam.lens.cam_origin[0], + cam.lens.cam_origin[1], + cam.lens.cam_origin[2], + cam.lens.focal_x, + cam.lens.focal_y, + cam.lens.fov, + cam.lens.alpha, + ) + cam_video_file = cam.video_file + transform_matrix = create_transform_matrix(focal_x, focal_y, cam_x, cam_y, skew_factor) + + for item_id, vals in metadata_results.items(): + world_coords, timestamps = vals + # print("timestamps are", timestamps) + world_coords = reformat_fetched_world_coords(world_coords) + cam_coords = world_to_pixel(world_coords, transform_matrix) + + vid_times = convert_datetime_to_frame_num(start_time, timestamps) + # print(vid_times) + + vid_fname = "./output/" + cam.metadata_id + item_id + ".mp4" + # print(vid_fname) + if boxed: + get_video_box(vid_fname, cam_video_file, cam_coords, vid_times) + else: + get_video_roi(vid_fname, cam_video_file, cam_coords, vid_times) + video_files.append(vid_fname) + print("output video files", ",".join(video_files)) + return video_files + + +def compile_lambda(pred): + s = uncompyle6.deparse_code2str(pred.__code__, out=open(os.devnull, "w")) + tree = ast.parse(s) + # print(pred.__code__) + # print(s) + # apprint(tree) + subtree = tree.body[0] + assert isinstance(subtree, ast.Return) + + x_range = [] + y_range = [] + + if isinstance(subtree.value, ast.BoolOp): + left_node = subtree.value.values[0] + right_node = subtree.value.values[1] + + # parse left + if isinstance(left_node, ast.Compare): + cmp_node = left_node + left = cmp_node.left + ops = cmp_node.ops + comparators = cmp_node.comparators + + if ( + len(comparators) == 2 + and isinstance(comparators[0], ast.Attribute) + and comparators[0].attr == "x" + ): + if isinstance(left, ast.BinOp): + if isinstance(left.left, ast.Attribute) and left.left.attr == "x": + if isinstance(left.op, ast.Sub): + assert isinstance(left.right, ast.Num) + x_range.append(-left.right.n) + elif isinstance(left.op, ast.Add): + assert isinstance(left.right, ast.Num) + x_range.append(left.right.n) + + if isinstance(comparators[-1], ast.BinOp): + right = comparators[-1] + if isinstance(right.left, ast.Attribute) and right.left.attr == "x": + if isinstance(right.op, ast.Sub): + assert isinstance(right.right, ast.Num) + x_range.append(-right.right.n) + elif isinstance(right.op, ast.Add): + assert isinstance(right.right, ast.Num) + x_range.append(right.right.n) + + if isinstance(right_node, ast.Compare): + cmp_node = right_node + left = cmp_node.left + ops = cmp_node.ops + comparators = cmp_node.comparators + + if ( + len(comparators) == 2 + and isinstance(comparators[0], ast.Attribute) + and comparators[0].attr == "y" + ): + if isinstance(left, ast.BinOp): + if isinstance(left.left, ast.Attribute) and left.left.attr == "y": + if isinstance(left.op, ast.Sub): + assert isinstance(left.right, ast.Num) + y_range.append(-left.right.n) + elif isinstance(left.op, ast.Add): + assert isinstance(left.right, ast.Num) + y_range.append(left.right.n) + + if isinstance(comparators[-1], ast.BinOp): + right = comparators[-1] + if isinstance(right.left, ast.Attribute) and right.left.attr == "y": + if isinstance(right.op, ast.Sub): + assert isinstance(right.right, ast.Num) + y_range.append(-right.right.n) + elif isinstance(right.op, ast.Add): + assert isinstance(right.right, ast.Num) + y_range.append(right.right.n) + return x_range, y_range diff --git a/apperception/new_world.py b/apperception/new_world.py new file mode 100644 index 00000000..770d2847 --- /dev/null +++ b/apperception/new_world.py @@ -0,0 +1,627 @@ +from __future__ import annotations + +import datetime +import glob +import inspect +import uuid +from collections.abc import Iterable +from enum import IntEnum +from os import makedirs, path +from pyclbr import Function +from typing import Any, Dict, List, Optional, Set, Tuple + +import cv2 +import dill as pickle +import matplotlib +import matplotlib.pyplot as plt +import numpy as np +import yaml +from bounding_box import WHOLE_FRAME, BoundingBox +from lens import Lens +from new_db import Database +from point import Point +from video_context import Camera + +from apperception.new_util import compile_lambda + +matplotlib.use("Qt5Agg") +print("get backend", matplotlib.get_backend()) + +makedirs("./.apperception_cache", exist_ok=True) + + +class Type(IntEnum): + # query type: for example, if we call get_cam(), and we execute the commands from root. when we encounter + # recognize(), we should not execute it because the inserted object must not be in the final result. we use enum + # type to determine whether we should execute this node + CAM, BBOX, TRAJ = 0, 1, 2 + + +BASE_VOLUME_QUERY_TEXT = "STBOX Z(({x1}, {y1}, {z1}),({x2}, {y2}, {z2}))" + + +class World: + # all worlds share a db instance + db = Database(reset=True) + camera_nodes: Dict[str, Camera] = {} + + _parent: Optional[World] + _name: str + # TODO: Fix _fn typing: (World, *Any, **Any) -> Query | str? | None + _fn: Any + _kwargs: dict[str, Any] + _done: bool + _world_id: str + _timestamp: datetime.datetime + _types: set[Type] + _materialized: bool + + def __init__( + self, + world_id: str, + timestamp: datetime.datetime, + name: str = None, + parent: World = None, + fn: Any = None, + kwargs: dict[str, Any] = None, + done: bool = False, + types: Set[Type] = None, + materialized: bool = False, + ): + self._parent = parent + self._name = "" if name is None else name + self._fn = None if fn is None else (getattr(self.db, fn) if isinstance(fn, str) else fn) + self._kwargs = {} if kwargs is None else kwargs + self._done = done # update node + self._world_id = world_id + self._timestamp = timestamp + self._types = set() if types is None else types + self._materialized = materialized + + def overlay_trajectory(self, cam_id, trajectory): + matplotlib.use( + "Qt5Agg" + ) # FIXME: matplotlib backend is agg here (should be qt5agg). Why is it overwritten? + print("get backend", matplotlib.get_backend()) + camera = World.camera_nodes[cam_id] + video_file = camera.video_file + for traj in trajectory: + current_trajectory = np.asarray(traj[0]) + frame_points = camera.lens.world_to_pixels(current_trajectory.T).T + vs = cv2.VideoCapture(video_file) + frame = vs.read() + frame = cv2.cvtColor(frame[1], cv2.COLOR_BGR2RGB) + for point in frame_points.tolist(): + cv2.circle(frame, tuple([int(point[0]), int(point[1])]), 3, (255, 0, 0)) + plt.figure() + plt.imshow(frame) + plt.show() + + def select_intersection_of_interest_or_use_default(self, cam_id, default=True): + camera = self.camera_nodes[cam_id] + video_file = camera.video_file + if default: + x1, y1, z1 = 0.01082532, 2.59647246, 0 + x2, y2, z2 = 3.01034039, 3.35985782, 2 + else: + vs = cv2.VideoCapture(video_file) + frame = vs.read() + frame = frame[1] + cv2.namedWindow("Frame", cv2.WINDOW_NORMAL) + cv2.resizeWindow("Frame", 384, 216) + initBB = cv2.selectROI("Frame", frame, fromCenter=False) + print(initBB) + cv2.destroyAllWindows() + print("world coordinate #1") + tl = camera.lens.pixel_to_world(initBB[:2], 1) + print(tl) + x1, y1, z1 = tl + print("world coordinate #2") + br = camera.lens.pixel_to_world((initBB[0] + initBB[2], initBB[1] + initBB[3]), 1) + print(br) + x2, y2, z2 = br + return BASE_VOLUME_QUERY_TEXT.format(x1=x1, y1=y1, z1=0, x2=x2, y2=y2, z2=2) + + def select_by_range(self, cam_id, x_range: Tuple[float, float], z_range: Tuple[float, float]): + camera = self.camera_nodes[cam_id] + x, _, z = camera.point.coordinate + + x_min = x + x_range[0] + x_max = x + x_range[1] + z_min = z + z_range[0] + z_max = z + z_range[1] + + return BASE_VOLUME_QUERY_TEXT.format( + x1=x_min, y1=float("-inf"), z1=z_min, x2=x_max, y2=float("inf"), z2=z_max + ) + + def recognize(self, cam_id: str, recognition_area: BoundingBox = WHOLE_FRAME): + assert cam_id in World.camera_nodes + + camera_node = World.camera_nodes[cam_id] + node1 = self._insert_bbox_traj(camera_node=camera_node, recognition_area=recognition_area) + node2 = node1._retrieve_bbox(world_id=node1._world_id) + node3 = node2._retrieve_traj(world_id=node1._world_id) + return node3 + + def get_video(self, cam_ids: List[str] = [], boxed: bool = False): + return derive_world( + self, + {Type.TRAJ}, + self.db.get_video, + cams=[World.camera_nodes[cam_id] for cam_id in cam_ids], + boxed=boxed, + )._execute_from_root(Type.TRAJ) + + def get_bbox(self): + return derive_world( + self, + {Type.BBOX}, + self.db.get_bbox, + )._execute_from_root(Type.BBOX) + + def get_traj(self): + return derive_world( + self, + {Type.TRAJ}, + self.db.get_traj, + )._execute_from_root(Type.TRAJ) + + def get_traj_key(self): + return derive_world( + self, + {Type.TRAJ}, + self.db.get_traj_key, + )._execute_from_root(Type.TRAJ) + + def get_headings(self): + # TODO: Optimize operations with NumPy if possible + trajectories = self.get_traj() + headings = [] + for traj in trajectories: + traj = traj[0] + heading = [None] + for j in range(1, len(traj)): + prev_pos = traj[j - 1] + current_pos = traj[j] + heading.append(0) + if current_pos[1] != prev_pos[1]: + heading[j] = np.arctan2( + current_pos[1] - prev_pos[1], current_pos[0] - prev_pos[0] + ) + heading[j] *= 180 / np.pi # convert to degrees from radian + heading[j] = ( + heading[j] + 360 + ) % 360 # converting such that all headings are positive + headings.append(heading) + return headings + + def get_distance(self, start: float, end: float): + return derive_world( + self, + {Type.TRAJ}, + self.db.get_distance, + start=str(self.db.start_time + datetime.timedelta(seconds=start)), + end=str(self.db.start_time + datetime.timedelta(seconds=end)), + )._execute_from_root(Type.TRAJ) + + def get_speed(self, start, end): + return derive_world( + self, + {Type.TRAJ}, + self.db.get_speed, + start=str(self.db.start_time + datetime.timedelta(seconds=start)), + end=str(self.db.start_time + datetime.timedelta(seconds=end)), + )._execute_from_root(Type.TRAJ) + + def filter_traj_type(self, object_type: str): + return derive_world(self, {Type.TRAJ}, self.db.filter_traj_type, object_type=object_type) + + def filter_traj_volume(self, volume: str): + return derive_world(self, {Type.TRAJ}, self.db.filter_traj_volume, volume=volume) + + def filter_traj_heading(self, lessThan=float("inf"), greaterThan=float("-inf")): + return derive_world( + self, + {Type.TRAJ}, + self.db.filter_traj_heading, + lessThan=lessThan, + greaterThan=greaterThan, + ) + + def filter_relative_to_type( + self, + x_range: Tuple[float, float], + y_range: Tuple[float, float], + z_range: Tuple[float, float], + type: str, + ): + return derive_world( + self, + {Type.TRAJ}, + self.db.filter_relative_to_type, + x_range=x_range, + y_range=y_range, + z_range=z_range, + type=type, + ) + + def filter_pred_relative_to_type(self, pred: Function): + x_range, y_range = compile_lambda(pred) + + return derive_world( + self, + {Type.TRAJ}, + self.db.filter_relative_to_type, + x_range=x_range, + y_range=y_range, + z_range=[float(-(2**31)), float(2**31)], + type="camera", + ) + + def add_camera( + self, + cam_id: str, + location: Point, + ratio: float, + video_file: str, + metadata_identifier: str, + lens: Lens, + ): + """ + 1. For update method, we create two nodes: the first node will write to the db, and the second node will retrieve from the db + 2. For the write node, never double write. (so we use done flag) + ... -> [write] -> [retrive] -> ... + """ + # TODO: should let the user add a Camera instead of its fields + camera_node = Camera(cam_id, location, ratio, video_file, metadata_identifier, lens) + World.camera_nodes[cam_id] = camera_node + + node1 = self._insert_camera(camera_node=camera_node) + node2 = node1._retrieve_camera(world_id=node1._world_id) + return node2 + + def interval(self, start, end): + return derive_world( + self, + {Type.BBOX}, + self.db.interval, + start=str(self.db.start_time + datetime.timedelta(seconds=start)), + end=str(self.db.start_time + datetime.timedelta(seconds=end)), + ) + + def add_properties(self, cam_id: str, properties: Any, property_type: str, new_prop): + # TODO: Should we add this to DB instead of the global object? + self.camera_nodes[cam_id].add_property(properties, property_type, new_prop) + + def predicate(self, condition: str): + return derive_world( + self, + {Type.CAM}, + self.db.filter_cam, + condition=condition, + ) + + def get_len(self): + return derive_world( + self, + {Type.CAM}, + self.db.get_len, + )._execute_from_root(Type.CAM) + + def get_camera(self): + return derive_world( + self, + {Type.CAM}, + self.db.get_cam, + )._execute_from_root(Type.CAM) + + def get_bbox_geo(self): + return derive_world( + self, + {Type.BBOX}, + self.db.get_bbox_geo, + )._execute_from_root(Type.BBOX) + + def get_time(self): + return derive_world( + self, + {Type.BBOX}, + self.db.get_time, + )._execute_from_root(Type.BBOX) + + def _insert_camera(self, camera_node: Camera): + return derive_world( + self, + {Type.CAM}, + self.db.insert_cam, + # does not pass in world_id because we want to use the world_id of the deriving world + camera_node=camera_node, + ) + + def _retrieve_camera(self, world_id: str): + return derive_world( + self, + {Type.CAM}, + self.db.retrieve_cam, + world_id=world_id, + ) + + def _insert_bbox_traj(self, camera_node: Camera, recognition_area: BoundingBox): + return derive_world( + self, + {Type.TRAJ, Type.BBOX}, + self.db.insert_bbox_traj, + # does not pass in world_id because we want to use the world_id of the deriving world + camera_node=camera_node, + recognition_area=recognition_area, + ) + + def _retrieve_bbox(self, world_id: str): + return derive_world(self, {Type.BBOX}, self.db.retrieve_bbox, world_id=world_id) + + def _retrieve_traj(self, world_id: str): + return derive_world(self, {Type.TRAJ}, self.db.retrieve_traj, world_id=world_id) + + def _execute_from_root(self, type: Type): + nodes: list[World] = [] + curr: Optional[World] = self + res = None + query = "" + + # collect all the nodes til the root + while curr: + nodes.append(curr) + curr = curr._parent + + # execute the nodes from the root + for node in nodes[::-1]: + # root + if node.fn is None: + continue + # if different type => pass + if type not in node.types: + continue + # treat update method differently + elif node.fn == self.db.insert_cam or node.fn == self.db.insert_bbox_traj: + if not node.done: + node._execute() + node._done = True + node._update_log_file() + else: + # print(query) + query = node._execute(query=query) + + res = query + return res + + def _execute(self, **kwargs): + fn_spec = inspect.getfullargspec(self._fn) + if "world_id" in fn_spec.args or fn_spec.varkw is not None: + return self._fn(**{"world_id": self._world_id, **self._kwargs, **kwargs}) + return self._fn(**{**self._kwargs, **kwargs}) + + def _print_lineage(self): + curr = self + while curr: + print(curr) + curr = curr._parent + + def __str__(self): + return ( + f"fn={self._fn}\nkwargs={self._kwargs}\ndone={self._done}\nworld_id={self._world_id}\n" + ) + + @property + def filename(self): + return filename(self._timestamp, self._world_id, self._name) + + @property + def world_id(self): + return self._world_id + + @property + def timestamp(self): + return self._timestamp + + @property + def parent(self): + return self._parent + + @property + def name(self): + return self._name + + @property + def fn(self): + return self._fn + + @property + def kwargs(self): + return self._kwargs + + @property + def done(self): + return self._done + + @property + def types(self): + return self._types + + @property + def materialized(self): + return self._materialized + + def _update_log_file(self): + with open(self.filename, "r") as f: + children = yaml.safe_load(f).get("children_filenames", None) + with open(self.filename, "w") as f: + f.write( + yaml.safe_dump( + { + **({} if self._parent is None else {"parent": self._parent.filename}), + **({} if self._types == set() else {"types": set(map(int, self._types))}), + **({} if self._fn is None else {"fn": self._fn.__name__}), + **({} if self._kwargs == {} else {"kwargs": pickle.dumps(self._kwargs)}), + **({} if not self._done else {"done": self._done}), + **({} if not self._materialized else {"materialized": self._materialized}), + **({} if children is None else {"children_filenames": children}), + } + ) + ) + + +def empty_world(name: str) -> World: + matched_files = list( + filter(path.isfile, glob.glob(f"./.apperception_cache/*_*_{name}.ap.yaml")) + ) + if len(matched_files): + return _empty_world_from_file(matched_files[0]) + return _empty_world(name) + + +def _empty_world_from_file(log_file: str) -> World: + with open(log_file, "r") as f: + content = yaml.safe_load(f) + if "children_filenames" in content: + del content["children_filenames"] + return World(*split_filename(log_file), **content) + + +def _empty_world(name: str) -> World: + world_id = str(uuid.uuid4()) + timestamp = datetime.datetime.utcnow() + log_file = filename(timestamp, world_id, name) + with open(log_file, "w") as f: + f.write(yaml.safe_dump({})) + return World(world_id, timestamp, name) + + +def derive_world(parent: World, types: set[Type], fn: Any, **kwargs) -> World: + world = _derive_world_from_file(parent, types, fn, **kwargs) + if world is not None: + return world + return _derive_world(parent, types, fn, **kwargs) + + +def _derive_world(parent: World, types: set[Type], fn: Any, **kwargs) -> World: + world_id = str(uuid.uuid4()) + timestamp = datetime.datetime.utcnow() + log_file = filename(timestamp, world_id) + + with open(parent.filename, "r") as pf: + content = yaml.safe_load(pf) + with open(parent.filename, "w") as pf: + content["children_filenames"] = content.get("children_filenames", set()) + content["children_filenames"].add(log_file) + pf.write(yaml.safe_dump(content)) + + with open(log_file, "w") as f: + f.write( + yaml.safe_dump( + { + "fn": fn.__name__, + "kwargs": pickle.dumps(kwargs), + "parent": parent.filename, + "types": set(map(int, types)), + } + ) + ) + + return World( + world_id, + timestamp, + fn=fn, + kwargs=kwargs, + parent=parent, + types=types, + ) + + +def _derive_world_from_file(parent: World, types: set[Type], fn: Any, **kwargs) -> Optional[World]: + with open(parent.filename, "r") as f: + sibling_filenames: Iterable[str] = yaml.safe_load(f).get("children_filenames", []) + + for sibling_filename in sibling_filenames: + with open(sibling_filename, "r") as sf: + sibling_content = yaml.safe_load(sf) + + if op_matched(sibling_content, types, fn, kwargs): + return World( + *split_filename(sibling_filename), + parent=parent, + **format_content(sibling_content), + ) + + return None + + +def from_file(filename: str) -> World: + with open(filename, "r") as f: + content = yaml.safe_load(f) + + parent_filename = content.get("parent_filename", None) + if parent_filename is None: + parent = None + else: + parent = from_file(parent_filename) + + return World(*split_filename(filename), parent=parent, **format_content(content)) + + +def filename(timestamp: datetime.datetime, world_id: str, name: str = ""): + return f".apperception_cache/{str(timestamp).replace(':', ';')}_{world_id}_{name}.ap.yaml" + + +def split_filename(filename: str) -> Tuple[str, datetime.datetime, str]: + filename = filename.replace("\\", "/") + timestamp_str, world_id, name = filename[: -len(".ap.yaml")].split("/")[-1].split("_", 2) + return world_id, datetime.datetime.fromisoformat(timestamp_str.replace(";", ":")), name + + +DUMPED_EMPTY_DICT = pickle.dumps({}) + + +def double_equal(a, b): + return a == b + + +def op_matched( + file_content: dict[str, Any], + types: set[Type], + fn: Any, + kwargs: dict[str, Any] = None, +) -> bool: + f_fn: str | None = file_content.get("fn", None) + f_types: set[int] = file_content.get("types", set()) + + if f_fn != fn.__name__ or f_types != set(map(int, types)): + return False + + kwargs = {} if kwargs is None else kwargs + f_kwargs: dict[str, Any] = pickle.loads(file_content.get("kwargs", DUMPED_EMPTY_DICT)) + + if len(f_kwargs) != len(kwargs): + return False + + cmps = fn.comparators if hasattr(fn, "comparators") else {} + return all( + key in f_kwargs and cmps.get(key, double_equal)(f_kwargs[key], kwargs[key]) + for key in kwargs + ) + + +def format_content(content: dict[str, Any]) -> dict[str, Any]: + if "types" in content: + content["types"] = set(map(Type, content["types"])) + + if "kwargs" in content: + content["kwargs"] = pickle.loads(content["kwargs"]) + + if "parent" in content: + del content["parent"] + + if "children_filenames" in content: + del content["children_filenames"] + + return content diff --git a/apperception/object_tracker_yolov5_deepsort.py b/apperception/object_tracker_yolov5_deepsort.py index ce20f858..b2763830 100644 --- a/apperception/object_tracker_yolov5_deepsort.py +++ b/apperception/object_tracker_yolov5_deepsort.py @@ -64,7 +64,7 @@ def detect(opt: YoloV5Opt): cfg.merge_from_file(opt.config_deepsort) attempt_download(deep_sort_weights, repo="mikel-brostrom/Yolov5_DeepSort_Pytorch") deepsort = DeepSort( - cfg.DEEPSORT.REID_CKPT, + deep_sort_weights, max_dist=cfg.DEEPSORT.MAX_DIST, min_confidence=cfg.DEEPSORT.MIN_CONFIDENCE, max_iou_distance=cfg.DEEPSORT.MAX_IOU_DISTANCE, @@ -92,30 +92,30 @@ def detect(opt: YoloV5Opt): names = model.module.names if hasattr(model, "module") else model.names # Run inference - if device.type != "cpu": - _, img, _, _ = dataset[0] - h, w = img.shape[1:] - - # crop image - x1, y1, x2, y2 = [ - int(v / 100.0) - for v in [ - w * crop.x1, - h * crop.y1, - w * crop.x2, - h * crop.y2, - ] - ] - - img = img[:, y1:y2, x1:x2] - model( - torch.zeros(1, 3, img.shape[1], img.shape[2]) - .to(device) - .type_as(next(model.parameters())) - ) # run once + # if device.type != "cpu": + # _, img, _, _ = dataset[0] + # h, w = img.shape[1:] + + # # crop image + # x1, y1, x2, y2 = [ + # int(v / 100.0) + # for v in [ + # w * crop.x1, + # h * crop.y1, + # w * crop.x2, + # h * crop.y2, + # ] + # ] + + # img = img[:, y1:y2, x1:x2] + # model( + # torch.zeros(1, 3, img.shape[1], img.shape[2]) + # .to(device) + # .type_as(next(model.parameters())) + # ) # run once formatted_result: Dict[str, TrackedObject] = {} - for frame_idx, (_, img, im0s, _) in enumerate(dataset): + for frame_idx, (_, img, im0s, _, _) in enumerate(dataset): h, w = img.shape[1:] # crop image @@ -152,7 +152,7 @@ def detect(opt: YoloV5Opt): continue # add padding from cropped frame - det[:, :4] += torch.tensor([[x1, y1, x1, y1]]) + det[:, :4] += torch.tensor([[x1, y1, x1, y1]]).to(device) # Rescale boxes from img_size to im0 size det[:, :4] = scale_coords( diff --git a/apperception/point.py b/apperception/point.py index 01691864..70c20d5e 100644 --- a/apperception/point.py +++ b/apperception/point.py @@ -1,20 +1,11 @@ +from dataclasses import dataclass +from typing import Tuple + + +@dataclass class Point: - def __init__( - self, - point_id: str, - object_id: str, - x: float, - y: float, - z: float, - time: float, - point_type: str, - ): - """ - Initializes an Point given coordinates, time, type and associated point ID - and object ID. - """ - self.point_id = point_id - self.object_id = object_id - self.coordinate = (x, y, z) - self.time = time - self.point_type = point_type + point_id: str + object_id: str + coordinate: Tuple[float, float, float] + time: float + point_type: str diff --git a/apperception/scenic_benchmark.py b/apperception/scenic_benchmark.py new file mode 100644 index 00000000..86ef2f94 --- /dev/null +++ b/apperception/scenic_benchmark.py @@ -0,0 +1,86 @@ +import lens +import point +from new_compute_lib import compute_heading +from new_world import empty_world + +# Let's define some attribute for constructing the world first +name = "traffic_scene" # world name +units = "metrics" # world units +video_file = "./amber_videos/traffic-scene-mini.mp4" # example video file +lens_attrs = {"fov": 120, "cam_origin": (0, 0, 0), "skew_factor": 0} +point_attrs = {"p_id": "p1", "cam_id": "cam1", "x": 0, "y": 0, "z": 0, "time": None, "type": "pos"} +camera_attrs = {"ratio": 0.5} +fps = 30 + +# 1. define a world +traffic_world = empty_world(name="my-world") + +# 2. construct a camera +fov, res, cam_origin, skew_factor = ( + lens_attrs["fov"], + [1280, 720], + lens_attrs["cam_origin"], + lens_attrs["skew_factor"], +) +cam_lens = lens.PinholeLens(res, cam_origin, fov, skew_factor) + +pt_id, cam_id, x, y, z, time, pt_type = ( + point_attrs["p_id"], + point_attrs["cam_id"], + point_attrs["x"], + point_attrs["y"], + point_attrs["z"], + point_attrs["time"], + point_attrs["type"], +) +location = point.Point(pt_id, cam_id, (x, y, z), time, pt_type) + +ratio = camera_attrs["ratio"] + +# ingest the camera into the world +traffic_world = traffic_world.add_camera( + cam_id=cam_id, + location=location, + ratio=ratio, + video_file=video_file, + metadata_identifier=name + "_" + cam_id, + lens=cam_lens, +) + +# Call execute on the world to run the detection algorithm and save the real data to the database +recognized_world = traffic_world.recognize(cam_id) + +# Use case #1 + +volume = traffic_world.select_by_range( + cam_id=cam_id, x_range=(0.01082532, 3.01034039), z_range=(0, 2) +) + + +filtered_world = recognized_world.filter_traj_type("car").filter_traj_volume(volume) +filtered_ids = filtered_world.get_traj_key() +print("filtered_ids are", filtered_ids) + +trajectory = filtered_world.get_traj() +print("trajectories are", trajectory) + +# Use case #2 + +volume = traffic_world.select_by_range( + cam_id=cam_id, x_range=(0.01082532, 3.01034039), z_range=(0, 2) +) + + +filtered_world = ( + recognized_world.filter_traj_type("car") + .filter_traj_volume(volume) + .filter_traj_heading(lessThan=8.5, greaterThan=-7.5) +) +filtered_ids = filtered_world.get_traj_key() +print("filtered_ids are", filtered_ids) + +trajectory = filtered_world.get_traj() +print("trajectories are", trajectory) + +heading = compute_heading(trajectory) +print("heading are", heading) diff --git a/apperception/test_parse_scenic_pred.py b/apperception/test_parse_scenic_pred.py new file mode 100644 index 00000000..c027653c --- /dev/null +++ b/apperception/test_parse_scenic_pred.py @@ -0,0 +1,95 @@ +import ast +import os + +import uncompyle6 +from astpretty import pprint as apprint + + +def main(): + def pred(obj): + return (cam.x - 10) <= obj.x <= (cam.x + 10) and (cam.y - 15) <= obj.y <= (cam.y + 70) + + s = uncompyle6.deparse_code2str(pred.__code__, out=open(os.devnull, "w")) + tree = ast.parse(s) + # print(pred.__code__) + # print(s) + # apprint(tree) + subtree = tree.body[0] + assert isinstance(subtree, ast.Return) + + apprint(subtree) + + x_range = [] + y_range = [] + + if isinstance(subtree.value, ast.BoolOp): + left_node = subtree.value.values[0] + right_node = subtree.value.values[1] + + # parse left + if isinstance(left_node, ast.Compare): + cmp_node = left_node + left = cmp_node.left + ops = cmp_node.ops + comparators = cmp_node.comparators + + if ( + len(comparators) == 2 + and isinstance(comparators[0], ast.Attribute) + and comparators[0].attr == "x" + ): + if isinstance(left, ast.BinOp): + if isinstance(left.left, ast.Attribute) and left.left.attr == "x": + if isinstance(left.op, ast.Sub): + assert isinstance(left.right, ast.Num) + x_range.append(-left.right.n) + elif isinstance(left.op, ast.Add): + assert isinstance(left.right, ast.Num) + x_range.append(left.right.n) + + if isinstance(comparators[-1], ast.BinOp): + right = comparators[-1] + if isinstance(right.left, ast.Attribute) and right.left.attr == "x": + if isinstance(right.op, ast.Sub): + assert isinstance(right.right, ast.Num) + x_range.append(-right.right.n) + elif isinstance(right.op, ast.Add): + assert isinstance(right.right, ast.Num) + x_range.append(right.right.n) + + if isinstance(right_node, ast.Compare): + cmp_node = right_node + left = cmp_node.left + ops = cmp_node.ops + comparators = cmp_node.comparators + + if ( + len(comparators) == 2 + and isinstance(comparators[0], ast.Attribute) + and comparators[0].attr == "y" + ): + if isinstance(left, ast.BinOp): + if isinstance(left.left, ast.Attribute) and left.left.attr == "y": + if isinstance(left.op, ast.Sub): + assert isinstance(left.right, ast.Num) + y_range.append(-left.right.n) + elif isinstance(left.op, ast.Add): + assert isinstance(left.right, ast.Num) + y_range.append(left.right.n) + + if isinstance(comparators[-1], ast.BinOp): + right = comparators[-1] + if isinstance(right.left, ast.Attribute) and right.left.attr == "y": + if isinstance(right.op, ast.Sub): + assert isinstance(right.right, ast.Num) + y_range.append(-right.right.n) + elif isinstance(right.op, ast.Add): + assert isinstance(right.right, ast.Num) + y_range.append(right.right.n) + + print(x_range) + print(y_range) + + +if __name__ == "__main__": + main() diff --git a/apperception/video_context.py b/apperception/video_context.py index 89cb46c7..e17c8bc6 100644 --- a/apperception/video_context.py +++ b/apperception/video_context.py @@ -7,6 +7,7 @@ import psycopg2 +@dataclass class Camera: def __init__(self, scenic_scene_name): self.scenic_scene_name = scenic_scene_name diff --git a/apperception/video_util.py b/apperception/video_util.py index a08d777a..067dae4b 100644 --- a/apperception/video_util.py +++ b/apperception/video_util.py @@ -1,4 +1,21 @@ -from typing import Any, Dict +from __future__ import annotations + +import datetime +import random +from typing import TYPE_CHECKING, Any, Dict, List, Optional + +import cv2 +import numpy as np +from bounding_box import WHOLE_FRAME, BoundingBox +from lens import Lens +from tracker import Tracker +from typing_extensions import Literal + +if TYPE_CHECKING: + from object_tracker_yolov5_deepsort import TrackedObject + +# TODO: add more units +Units = Literal["metrics"] def video_data_to_tasm(video_file, metadata_id, t): @@ -33,7 +50,127 @@ def bound_height(y): t.add_bulk_metadata(metadata_info) -def create_or_insert_world_table(conn, name, units): +def convert_datetime_to_frame_num(start_time, date_times): + + return [(t.replace(tzinfo=None) - start_time).total_seconds() for t in date_times] + + +def get_video_roi(file_name, cam_video_file, rois, times): + """ + Get the region of interest from the video, based on bounding box points in + video coordinates. + + Args: + file_name: String of file name to save video as + rois: A list of bounding boxes + time_intervals: A list of time intervals of which frames + """ + + rois = np.array(rois).T + print(rois.shape) + len_x, len_y = np.max(rois.T[2] - rois.T[0]), np.max(rois.T[3] - rois.T[1]) + # len_x, len_y = np.max(rois.T[0][1] - rois.T[0][0]), np.max(rois.T[1][1] - rois.T[1][0]) + + len_x = int(round(len_x)) + len_y = int(round(len_y)) + # print(len_x) + # print(len_y) + vid_writer = cv2.VideoWriter( + file_name, cv2.VideoWriter_fourcc("m", "p", "4", "v"), 30, (len_x, len_y) + ) + # print("rois") + # print(rois) + start_time = int(times[0]) + cap = cv2.VideoCapture(cam_video_file) + frame_cnt = 0 + while cap.isOpened(): + # Capture frame-by-frame + ret, frame = cap.read() + if frame_cnt in times and ret: + i = frame_cnt - start_time + if i >= len(rois): + print("incorrect length:", len(rois)) + break + current_roi = rois[i] + + b_x, b_y, e_x, e_y = current_roi + b_x, b_y = max(0, b_x), max(0, b_y) + # e_x, e_y = current_roi[1] + e_x, e_y = max(0, e_x), max(0, e_y) + diff_y, diff_x = int(abs(e_y - b_y)), int(abs(e_x - b_x)) + pad_y = int((len_y - diff_y) // 2) + pad_x = int((len_x - diff_x) // 2) + + # print("padding") + # print(pad_y) + # print(pad_x) + roi_byte = frame[int(b_y) : int(e_y), int(b_x) : int(e_x), :] + + roi_byte = np.pad( + roi_byte, + pad_width=[ + (pad_y, len_y - diff_y - pad_y), + (pad_x, len_x - diff_x - pad_x), + (0, 0), + ], + ) + frame = cv2.cvtColor(roi_byte, cv2.COLOR_RGB2BGR) + + vid_writer.write(roi_byte) + frame_cnt += 1 + if not ret: + break + + vid_writer.release() + + +def get_video_box( + file_name: str, cam_video_file: str, rois: List[(int, int, int, int)], times: List[int] +): + """ + Get the frames of interest from the video, while boxing in the object at interest + with a box. + + Args: + file_name: String of file name to save video as + rois: A list of bounding boxes + time_intervals: A list of time intervals of which frames + """ + + rois = np.array(rois).T + print(rois.shape) + + cap = cv2.VideoCapture(cam_video_file) + width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)) + height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) + vid_writer = cv2.VideoWriter( + file_name, cv2.VideoWriter_fourcc("m", "p", "4", "v"), 30, (width, height) + ) + + start_time = int(times[0]) + frame_cnt = 0 + while cap.isOpened(): + # Capture frame-by-frame + ret, frame = cap.read() + if frame_cnt in times and ret: + i = frame_cnt - start_time + if i >= len(rois): + print("incorrect length:", len(rois)) + break + current_roi = rois[i] + + x1, y1, x2, y2 = current_roi + cv2.rectangle(frame, (int(x1), int(y1)), (int(x2), int(y2)), (255, 255, 0), 2) + + vid_writer.write(frame) + frame_cnt += 1 + if not ret: + break + + vid_writer.release() + + +def create_or_insert_world_table(conn, name, units: Units): # Creating a cursor object using the cursor() method cursor = conn.cursor() """ @@ -64,3 +201,374 @@ def insert_world(conn, name, units): print("New world inserted successfully........") # Insert the existing cameras of the current world into the camera table conn.commit() + + +def create_or_insert_camera_table(conn, world_name, camera): + """Create a camera table""" + # Creating a cursor object using the cursor() method + cursor = conn.cursor() + """ + Create and Populate A camera table with the given camera object. + """ + # Creating table with the first camera + sql = "\n".join( + [ + "CREATE TABLE IF NOT EXISTS Cameras(", + " cameraId TEXT,", + " worldId TEXT,", + " ratio real,", + " origin geometry,", + " focalpoints geometry,", + " fov INTEGER,", + " skev_factor real,", + " width integer,", + " height integer", + ");", + ] + ) + cursor.execute(sql) + print("Camera Table created successfully........") + insert_camera(conn, world_name, camera) + return sql + + +def insert_camera(conn, world_name, camera_node): + """Helper function to insert the camera""" + # Creating a cursor object using the cursor() method + cursor = conn.cursor() + lens = camera_node.lens + focal_x = str(lens.focal_x) + focal_y = str(lens.focal_y) + cam_x, cam_y, cam_z = str(lens.cam_origin[0]), str(lens.cam_origin[1]), str(lens.cam_origin[2]) + width, height = camera_node.dimension + cursor.execute( + """INSERT INTO Cameras (cameraId, worldId, ratio, origin, focalpoints, fov, skev_factor, width, height) """ + + """VALUES (\'%s\', \'%s\', %f, \'POINT Z (%s %s %s)\', \'POINT(%s %s)\', %s, %f, %d, %d);""" + % ( + camera_node.cam_id, + world_name, + camera_node.ratio, + cam_x, + cam_y, + cam_z, + focal_x, + focal_y, + lens.fov, + lens.alpha, + width, + height, + ) + ) + print("New camera inserted successfully.........") + conn.commit() + + +def get_video_dimension(video_file: str): + vid: cv2.VideoCapture = cv2.VideoCapture(video_file) + width = vid.get(cv2.CAP_PROP_FRAME_WIDTH) + height = vid.get(cv2.CAP_PROP_FRAME_HEIGHT) + # width and height are floats + return (int(width), int(height)) + + +def recognize( + video_file: str, + recog_algo: str = "", + tracker_type: str = "default", + customized_tracker: Optional[Tracker] = None, + recognition_area: BoundingBox = WHOLE_FRAME, +): + """Default object recognition (YOLOv5)""" + from object_tracker_yolov4_deepsort import yolov4_deepsort_video_track + from object_tracker_yolov5_deepsort import (YoloV5Opt, + yolov5_deepsort_video_track) + + # recognition = item.ItemRecognition(recog_algo = recog_algo, tracker_type = tracker_type, customized_tracker = customized_tracker) + # return recognition.video_item_recognize(video.byte_array) + if recognition_area.is_whole_frame(): + recognition_area = BoundingBox(0, 0, 100, 100) + if recog_algo == "yolov4": + return yolov4_deepsort_video_track(video_file, recognition_area) + else: + # use YoloV5 as default + return yolov5_deepsort_video_track(YoloV5Opt(video_file, recognition_area=recognition_area)) + + +def add_recognized_objs( + conn: Any, + lens: Lens, + formatted_result: Dict[str, TrackedObject], + start_time: datetime.datetime, + world_id: str = "default", + properties: dict = {"color": {}}, + default_depth: bool = True, +): + # # TODO: move cleaning to apperception_benchmark.py + # clean_tables(conn) + for item_id in formatted_result: + object_type = formatted_result[item_id].object_type + recognized_bboxes = np.array( + [bbox.to_tuples() for bbox in formatted_result[item_id].bboxes] + ) + tracked_cnt = formatted_result[item_id].tracked_cnt + top_left = np.vstack((recognized_bboxes[:, 0, 0], recognized_bboxes[:, 0, 1])) + if default_depth: + top_left_depths = np.ones(len(recognized_bboxes)) + else: + top_left_depths = self.__get_depths_of_points( + recognized_bboxes[:, 0, 0], recognized_bboxes[:, 0, 1] + ) + top_left = lens.pixels_to_world(top_left, top_left_depths) + + # Convert bottom right coordinates to world coordinates + bottom_right = np.vstack((recognized_bboxes[:, 1, 0], recognized_bboxes[:, 1, 1])) + if default_depth: + bottom_right_depths = np.ones(len(tracked_cnt)) + else: + bottom_right_depths = self.__get_depths_of_points( + recognized_bboxes[:, 1, 0], recognized_bboxes[:, 1, 1] + ) + bottom_right = lens.pixels_to_world(bottom_right, bottom_right_depths) + + top_left = np.array(top_left.T) + bottom_right = np.array(bottom_right.T) + obj_traj = [] + for i in range(len(top_left)): + current_tl = top_left[i] + current_br = bottom_right[i] + obj_traj.append([current_tl.tolist(), current_br.tolist()]) + + bbox_to_postgres( + conn, + item_id, + object_type, + "default_color" if item_id not in properties["color"] else properties["color"][item_id], + random.uniform(-10, 10), # heading + start_time, + tracked_cnt, + obj_traj, + world_id, + type="yolov4", + ) + # bbox_to_tasm() + + +def convert_timestamps(start_time, timestamps): + """Helper function to convert the timestam to the timestamp formula pg-trajectory uses""" + return [str(start_time + datetime.timedelta(seconds=t)) for t in timestamps] + + +def bbox_to_data3d(bbox): + """Helper function to convert trajectory to centroids + Compute the center, x, y, z delta of the bbox + """ + tl, br = bbox + x_delta = (br[0] - tl[0]) / 2 + y_delta = (br[1] - tl[1]) / 2 + z_delta = (br[2] - tl[2]) / 2 + center = (tl[0] + x_delta, tl[1] + y_delta, tl[2] + z_delta) + + return center, x_delta, y_delta, z_delta + + +def bbox_to_postgres( + conn, + item_id, + object_type, + color, + heading, + start_time, + timestamps, + bboxes, + world_id="default", + type="yolov3", +): + """Insert bboxes to postgres""" + if type == "yolov3": + timestamps = range(timestamps) + + converted_bboxes = [bbox_to_data3d(bbox) for bbox in bboxes] + pairs = [] + deltas = [] + for meta_box in converted_bboxes: + pairs.append(meta_box[0]) + deltas.append(meta_box[1:]) + postgres_timestamps = convert_timestamps(start_time, timestamps) + create_or_insert_general_trajectory( + conn, item_id, object_type, color, heading, postgres_timestamps, bboxes, pairs, world_id + ) + print(f"{item_id} saved successfully") + + +def clean_tables(conn): + cursor = conn.cursor() + cursor.execute("DROP TABLE IF EXISTS General_Bbox;") + cursor.execute("DROP TABLE IF EXISTS Item_General_Trajectory;") + conn.commit() + + +def create_or_insert_general_trajectory( + conn, + item_id, + object_type, + color, + heading, + postgres_timestamps, + bboxes, + pairs, + world_id="default", +): + """Create general trajectory table""" + # Creating a cursor object using the cursor() method + cursor = conn.cursor() + """ + Create and Populate A Trajectory table using mobilityDB. + Now the timestamp matches, the starting time should be the meta data of the world + Then the timestamp should be the timestamp regarding the world starting time + """ + + # Creating table with the first item + create_itemtraj_sql = """CREATE TABLE IF NOT EXISTS Item_General_Trajectory( + itemId TEXT, + objectType TEXT, + color TEXT, + heading REAL, + trajCentroids tgeompoint, + largestBbox stbox, + worldId TEXT, + PRIMARY KEY (itemId) + );""" + cursor.execute(create_itemtraj_sql) + cursor.execute( + "CREATE INDEX IF NOT EXISTS traj_idx ON Item_General_Trajectory USING GiST(trajCentroids);" + ) + conn.commit() + # Creating table with the first item + create_bboxes_sql = """CREATE TABLE IF NOT EXISTS General_Bbox( + itemId TEXT, + trajBbox stbox, + worldId TEXT, + FOREIGN KEY(itemId) + REFERENCES Item_General_Trajectory(itemId) + );""" + cursor.execute(create_bboxes_sql) + cursor.execute("CREATE INDEX IF NOT EXISTS item_idx ON General_Bbox(itemId);") + cursor.execute("CREATE INDEX IF NOT EXISTS traj_bbox_idx ON General_Bbox USING GiST(trajBbox);") + conn.commit() + # Insert the trajectory of the first item + insert_general_trajectory( + conn, item_id, object_type, color, heading, postgres_timestamps, bboxes, pairs, world_id + ) + + +def insert_general_trajectory( + conn, + item_id, + object_type, + color, + heading, + postgres_timestamps, + bboxes, + pairs, + world_id="default", +): + """Insert general trajectory""" + # Creating a cursor object using the cursor() method + cursor = conn.cursor() + # Inserting bboxes into Bbox table + insert_bbox_trajectory = "" + insert_format = ( + "INSERT INTO General_Bbox (itemId, worldId, trajBbox) " + + "VALUES ('%s','%s'," % (item_id + "-" + world_id, world_id) + ) + # Insert the item_trajectory separately + insert_trajectory = ( + "INSERT INTO Item_General_Trajectory (itemId, worldId, objectType, color, heading, trajCentroids, largestBbox) " + + "VALUES ('%s', '%s', '%s', '%s', '%s', " + % (item_id + "-" + world_id, world_id, object_type, color, heading) + ) + traj_centroids = "'{" + min_ltx, min_lty, min_ltz, max_brx, max_bry, max_brz = ( + float("inf"), + float("inf"), + float("inf"), + float("-inf"), + float("-inf"), + float("-inf"), + ) + # max_ltx, max_lty, max_ltz, min_brx, min_bry, min_brz = float('-inf'), float('-inf'), float('-inf'), float('inf'), float('inf'), float('inf') + for i in range(len(postgres_timestamps)): + postgres_timestamp = postgres_timestamps[i] + # Insert bbox + # print(bboxes[i]) + tl, br = bboxes[i] + min_ltx, min_lty, min_ltz, max_brx, max_bry, max_brz = ( + min(tl[0], min_ltx), + min(tl[1], min_lty), + min(tl[2], min_ltz), + max(br[0], max_brx), + max(br[1], max_bry), + max(br[2], max_brz), + ) + # max_ltx, max_lty, max_ltz, min_brx, min_bry, min_brz = max(tl[0], max_ltx), max(tl[1], max_lty), max(tl[2], max_ltz),\ + # min(br[0], min_brx), min(br[1], min_bry), min(br[2], min_brz) + current_bbox_sql = "stbox 'STBOX ZT((%s, %s, %s, %s), (%s, %s, %s, %s))');" % ( + tl[0], + tl[1], + tl[2], + postgres_timestamp, + br[0], + br[1], + br[2], + postgres_timestamp, + ) + insert_bbox_trajectory += insert_format + current_bbox_sql + # Construct trajectory + current_point = pairs[i] + tg_pair_centroid = "POINT Z (%s %s %s)@%s," % ( + str(current_point[0]), + str(current_point[1]), + str(current_point[2]), + postgres_timestamp, + ) + traj_centroids += tg_pair_centroid + traj_centroids = traj_centroids[:-1] + traj_centroids += "}', " + insert_trajectory += traj_centroids + insert_trajectory += "stbox 'STBOX Z((%s, %s, %s)," % ( + min_ltx, + min_lty, + min_ltz, + ) + "(%s, %s, %s))'); " % (max_brx, max_bry, max_brz) + # print(insert_trajectory) + cursor.execute(insert_trajectory) + cursor.execute(insert_bbox_trajectory) + # Commit your changes in the database + conn.commit() + + +def merge_trajectory(item_id, new_postgres_timestamps, new_bboxes, new_pairs): + # Fetch the timestamps of the current trajectory from the database + # Filter out the already had timestamp from the new timestamps + # Construct the adding trajectory + # Calling the merge function of mobilitydb + # do the same thing for the bboxes + return + + +def fetch_camera(conn, world_id="default", cam_id=[]): + cursor = conn.cursor() + + if cam_id == []: + query = ( + """SELECT cameraId, ratio, ST_X(origin), ST_Y(origin), ST_Z(origin), ST_X(focalpoints), ST_Y(focalpoints), fov, skev_factor """ + + """FROM Cameras WHERE worldId = \'%s\';""" % world_id + ) + else: + query = ( + """SELECT cameraId, ratio, ST_X(origin), ST_Y(origin), ST_Z(origin), ST_X(focalpoints), ST_Y(focalpoints), fov, skev_factor """ + + """FROM Cameras WHERE cameraId IN (\'%s\') AND worldId = \'%s\';""" + % (",".join(cam_id), world_id) + ) + cursor.execute(query) + return cursor.fetchall() diff --git a/apperception/world.py b/apperception/world.py index f5eedbdc..dadd9ce8 100644 --- a/apperception/world.py +++ b/apperception/world.py @@ -48,6 +48,7 @@ def get_name(self): def get_units(self): return self.VideoContext.get_units() + # TODO: should be add_item / add_camera? def item(self, item_id, cam_id, item_type, location): new_context = copy.deepcopy(self) new_context.VideoContext.item(item_id, cam_id, item_type, location) diff --git a/apperception_example.ipynb b/apperception_example.ipynb index 12dae4be..c17210ea 100644 --- a/apperception_example.ipynb +++ b/apperception_example.ipynb @@ -125,27 +125,53 @@ "output_type": "stream", "name": "stdout", "text": [ - "Worlds Table created successfully........\n", - "New world inserted successfully........\n", - "Camera Table created successfully........\n", - "New camera inserted successfully.........\n", - "# of tracked items: 16\n", - "car-1 saved successfully\n", - "car-2 saved successfully\n", - "person-3 saved successfully\n", - "person-4 saved successfully\n", - "person-5 saved successfully\n", - "traffic light-6 saved successfully\n", - "traffic light-7 saved successfully\n", + "video 1/1 (1/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: " + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/Users/byronhsu/Desktop/apperception/env/lib/python3.8/site-packages/torch/functional.py:445: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:2157.)\n", + " return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "video 1/1 (2/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (3/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (4/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (5/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (6/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (7/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (8/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (9/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (10/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (11/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (12/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (13/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (14/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (15/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (16/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (17/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (18/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (19/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (20/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (21/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (22/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (23/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (24/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (25/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (26/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (27/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (28/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (29/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (30/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (31/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (32/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (33/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (34/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (35/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (36/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (37/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (38/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (39/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (40/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (41/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (42/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (43/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (44/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (45/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (46/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (47/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (48/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (49/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (50/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (51/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (52/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (53/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (54/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (55/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (56/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (57/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (58/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (59/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (60/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (61/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (62/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (63/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (64/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (65/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (66/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (67/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (68/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (69/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (70/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (71/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (72/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (73/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (74/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (75/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (76/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (77/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (78/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (79/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (80/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (81/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (82/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (83/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (84/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (85/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (86/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (87/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: " + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "video 1/1 (88/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (89/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (90/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (91/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (92/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (93/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (94/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (95/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (96/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (97/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (98/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (99/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (100/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (101/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (102/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (103/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (104/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (105/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (106/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (107/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (108/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (109/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (110/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (111/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (112/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (113/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (114/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (115/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (116/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (117/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (118/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (119/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (120/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (121/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (122/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: video 1/1 (123/123) /Users/byronhsu/Desktop/apperception/amber_videos/traffic-scene-shorter.mp4: car-1 saved successfully\n", + "person-2 saved successfully\n", + "traffic light-3 saved successfully\n", + "car-4 saved successfully\n", + "car-6 saved successfully\n", + "person-7 saved successfully\n", + "traffic light-8 saved successfully\n", + "person-9 saved successfully\n", "person-10 saved successfully\n", "person-11 saved successfully\n", - "car-12 saved successfully\n", - "person-13 saved successfully\n", - "car-15 saved successfully\n", - "car-17 saved successfully\n", - "person-21 saved successfully\n", + "traffic light-12 saved successfully\n", + "person-12 saved successfully\n", + "car-13 saved successfully\n", + "person-15 saved successfully\n", + "car-18 saved successfully\n", + "traffic light-10 saved successfully\n", + "traffic light-9 saved successfully\n", + "person-19 saved successfully\n", + "car-20 saved successfully\n", + "truck-20 saved successfully\n", + "bus-20 saved successfully\n", + "person-24 saved successfully\n", "person-25 saved successfully\n", - "person-26 saved successfully\n" + "traffic light-25 saved successfully\n", + "traffic light-11 saved successfully\n" ] } ], diff --git a/new_apperception_benchmark.ipynb b/new_apperception_benchmark.ipynb new file mode 100644 index 00000000..c721cb0f --- /dev/null +++ b/new_apperception_benchmark.ipynb @@ -0,0 +1,426 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "id": "49d6ebd1", + "metadata": {}, + "outputs": [], + "source": [ + "# Deleting the .apperception_cache if it exists, as to avoid DB conflict errors\n", + "import os\n", + "import shutil\n", + "\n", + "dirpath = os.path.join('.apperception_cache')\n", + "if os.path.exists(dirpath) and os.path.isdir(dirpath):\n", + " shutil.rmtree(dirpath)" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "04d5291c", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "get backend Qt5Agg\n" + ] + } + ], + "source": [ + "import sys\n", + "sys.path.append(os.path.join(os.getcwd(),\"apperception\"))\n", + "\n", + "### IMPORTS\n", + "import lens\n", + "import point\n", + "from new_world import empty_world\n", + "\n", + "# Let's define some attribute for constructing the world first\n", + "name = \"trafficScene\" # world name\n", + "units = \"metrics\" # world units\n", + "video_file = \"amber_videos/traffic-scene-shorter.mp4\" # example video file\n", + "lens_attrs = {\"fov\": 120, \"cam_origin\": (0, 0, 0), \"skew_factor\": 0}\n", + "point_attrs = {\"p_id\": \"p1\", \"cam_id\": \"cam1\", \"x\": 0, \"y\": 0, \"z\": 0, \"time\": None, \"type\": \"pos\"}\n", + "camera_attrs = {\"ratio\": 0.5}\n", + "fps = 30\n", + "\n", + "# 1. define a world\n", + "traffic_world = empty_world(name)\n", + "\n", + "# 2. construct a camera\n", + "fov, res, cam_origin, skew_factor = (\n", + " lens_attrs[\"fov\"],\n", + " [1280, 720],\n", + " lens_attrs[\"cam_origin\"],\n", + " lens_attrs[\"skew_factor\"],\n", + ")\n", + "cam_lens = lens.PinholeLens(res, cam_origin, fov, skew_factor)\n", + "\n", + "pt_id, cam_id, x, y, z, time, pt_type = (\n", + " point_attrs[\"p_id\"],\n", + " point_attrs[\"cam_id\"],\n", + " point_attrs[\"x\"],\n", + " point_attrs[\"y\"],\n", + " point_attrs[\"z\"],\n", + " point_attrs[\"time\"],\n", + " point_attrs[\"type\"],\n", + ")\n", + "location = point.Point(pt_id, cam_id, (x, y, z), time, pt_type)\n", + "\n", + "ratio = camera_attrs[\"ratio\"]\n", + "\n", + "# ingest the camera into the world\n", + "traffic_world = traffic_world.add_camera(\n", + " cam_id=cam_id,\n", + " location=location,\n", + " ratio=ratio,\n", + " video_file=video_file,\n", + " metadata_identifier=name + \"_\" + cam_id,\n", + " lens=cam_lens,\n", + ")\n", + "\n", + "# Call execute on the world to run the detection algorithm and save the real data to the database\n", + "recognized_world = traffic_world.recognize(cam_id)\n", + "\n", + "volume = traffic_world.select_intersection_of_interest_or_use_default(cam_id=cam_id)" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "5f1f65fd", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "cameras are [('cam1', 0.5, 0.0, 0.0, 0.0, 369.5041722813606, 207.84609690826534, 120, 0.0)]\n", + "lens are [(0.5, 0.0, 0.0, 0.0, 120, 0.0)]\n" + ] + } + ], + "source": [ + "cams = traffic_world.get_camera()\n", + "lens = traffic_world.get_len()\n", + "# ids = traffic_world.get_id()\n", + "print(\"cameras are\", cams)\n", + "print(\"lens are\", lens)\n", + "# print(\"ids are\", ids)" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "3bd024c9", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Loading weights from C:\\Users\\youse\\Desktop\\Research\\Apperception\\apperception\\apperception\\../yolov5-deepsort/deep_sort_pytorch/deep_sort/deep/checkpoint/ckpt.t7... Done!\n", + "YOLOv5 v6.0-159-gdb6ec66 torch 1.10.2+cu113 CUDA:0 (NVIDIA GeForce RTX 3060 Laptop GPU, 6144MiB)\n", + "\n", + "Fusing layers... \n", + "Model Summary: 213 layers, 7225885 parameters, 0 gradients\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "car-1 saved successfully\n", + "person-2 saved successfully\n", + "traffic light-3 saved successfully\n", + "car-4 saved successfully\n", + "car-6 saved successfully\n", + "person-7 saved successfully\n", + "traffic light-8 saved successfully\n", + "person-9 saved successfully\n", + "person-10 saved successfully\n", + "traffic light-12 saved successfully\n", + "person-12 saved successfully\n", + "person-13 saved successfully\n", + "car-14 saved successfully\n", + "person-16 saved successfully\n", + "bus-14 saved successfully\n", + "car-19 saved successfully\n", + "traffic light-10 saved successfully\n", + "traffic light-13 saved successfully\n", + "person-20 saved successfully\n", + "car-21 saved successfully\n", + "truck-21 saved successfully\n", + "bus-21 saved successfully\n", + "person-25 saved successfully\n", + "person-26 saved successfully\n", + "traffic light-26 saved successfully\n", + "traffic light-9 saved successfully\n", + "get_traj_key SELECT sq2.itemid FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM item_general_trajectory WHERE worldId='a29b573e-70a0-4b6c-b19d-4f453867ce4f') sq0 WHERE sq0.objecttype='car') sq1 WHERE overlap(sq1.largestBbox,'STBOX Z((0.01082532, 2.59647246, 0),(3.01034039, 3.35985782, 2))')) sq2\n", + "filtered_ids are [('car-1-a29b573e-70a0-4b6c-b19d-4f453867ce4f',), ('car-4-a29b573e-70a0-4b6c-b19d-4f453867ce4f',), ('car-14-a29b573e-70a0-4b6c-b19d-4f453867ce4f',), ('car-21-a29b573e-70a0-4b6c-b19d-4f453867ce4f',)]\n" + ] + } + ], + "source": [ + "# This piece of code is unsafe, and should not be run if not needed. \n", + "# It serves for test purposes when one recieves a \"dead kernel\" error.\n", + "# os.environ[\"KMP_DUPLICATE_LIB_OK\"]=\"TRUE\"\n", + "\n", + "filtered_world = recognized_world.filter_traj_type(\"car\").filter_traj_volume(volume).interval(0, fps * 3)\n", + "filtered_ids = filtered_world.get_traj_key()\n", + "print(\"filtered_ids are\", filtered_ids)" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "6e5d5bc6", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "get_traj SELECT asMFJSON(trajCentroids)::json->'coordinates' FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM item_general_trajectory WHERE worldId='a29b573e-70a0-4b6c-b19d-4f453867ce4f') sq0 WHERE sq0.objecttype='car') sq1 WHERE overlap(sq1.largestBbox,'STBOX Z((0.01082532, 2.59647246, 0),(3.01034039, 3.35985782, 2))')) as final\n", + "trajectories are [([[2.212424273730558, 2.574019950137081, 1], [2.212424273730558, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.212424273730558, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.213777438423971, 2.574019950137081, 1], [2.212424273730558, 2.574019950137081, 1], [2.212424273730558, 2.574019950137081, 1], [2.211071109037144, 2.571614324015457, 1], [2.211071109037144, 2.571614324015457, 1], [2.211071109037144, 2.571614324015457, 1], [2.211071109037144, 2.571614324015457, 1], [2.211071109037144, 2.569208697893834, 1], [2.209717944343731, 2.574019950137081, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.571614324015457, 1], [2.207011614956905, 2.569208697893834, 1], [2.208364779650318, 2.574019950137081, 1], [2.208364779650318, 2.571614324015457, 1], [2.208364779650318, 2.571614324015457, 1], [2.208364779650318, 2.574019950137081, 1], [2.208364779650318, 2.574019950137081, 1], [2.208364779650318, 2.574019950137081, 1], [2.211071109037144, 2.574019950137081, 1], [2.211071109037144, 2.574019950137081, 1], [2.212424273730558, 2.574019950137081, 1], [2.211071109037144, 2.574019950137081, 1], [2.211071109037144, 2.574019950137081, 1], [2.208364779650318, 2.574019950137081, 1], [2.208364779650318, 2.576425576258704, 1], [2.211071109037144, 2.576425576258704, 1], [2.211071109037144, 2.576425576258704, 1], [2.211071109037144, 2.574019950137081, 1], [2.211071109037144, 2.574019950137081, 1], [2.211071109037144, 2.574019950137081, 1], [2.209717944343731, 2.574019950137081, 1], [2.209717944343731, 2.574019950137081, 1], [2.208364779650318, 2.574019950137081, 1], [2.208364779650318, 2.574019950137081, 1], [2.208364779650318, 2.574019950137081, 1], [2.207011614956905, 2.574019950137081, 1], [2.208364779650318, 2.574019950137081, 1], [2.209717944343731, 2.571614324015457, 1], [2.211071109037144, 2.571614324015457, 1], [2.21377743842397, 2.569208697893834, 1], [2.211071109037144, 2.574019950137081, 1], [2.21377743842397, 2.569208697893834, 1], [2.212424273730557, 2.569208697893834, 1], [2.212424273730557, 2.564397445650587, 1], [2.21377743842397, 2.564397445650587, 1], [2.21377743842397, 2.55958619340734, 1], [2.211071109037144, 2.55958619340734, 1], [2.211071109037144, 2.566803071772211, 1], [2.211071109037144, 2.564397445650587, 1], [2.211071109037144, 2.564397445650587, 1], [2.211071109037144, 2.564397445650587, 1], [2.211071109037144, 2.564397445650587, 1], [2.211071109037144, 2.566803071772211, 1], [2.209717944343731, 2.569208697893834, 1], [2.208364779650318, 2.571614324015457, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.571614324015457, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.564397445650587, 1], [2.208364779650318, 2.564397445650587, 1], [2.208364779650318, 2.564397445650587, 1], [2.208364779650318, 2.569208697893834, 1], [2.208364779650318, 2.569208697893834, 1], [2.209717944343731, 2.569208697893834, 1], [2.209717944343731, 2.569208697893834, 1], [2.211071109037144, 2.564397445650587, 1], [2.211071109037144, 2.566803071772211, 1], [2.212424273730557, 2.566803071772211, 1], [2.212424273730557, 2.569208697893834, 1], [2.212424273730557, 2.566803071772211, 1], [2.211071109037144, 2.569208697893834, 1], [2.211071109037144, 2.569208697893834, 1], [2.211071109037144, 2.569208697893834, 1], [2.211071109037144, 2.569208697893834, 1], [2.211071109037144, 2.569208697893834, 1], [2.211071109037144, 2.569208697893834, 1], [2.209717944343731, 2.569208697893834, 1], [2.211071109037144, 2.569208697893834, 1], [2.209717944343731, 2.569208697893834, 1], [2.209717944343731, 2.574019950137081, 1], [2.211071109037144, 2.571614324015457, 1], [2.209717944343731, 2.569208697893834, 1], [2.211071109037144, 2.566803071772211, 1], [2.211071109037144, 2.566803071772211, 1], [2.211071109037144, 2.566803071772211, 1], [2.211071109037144, 2.569208697893834, 1], [2.211071109037144, 2.566803071772211, 1], [2.211071109037144, 2.566803071772211, 1], [2.209717944343731, 2.566803071772211, 1], [2.209717944343731, 2.569208697893834, 1], [2.209717944343731, 2.569208697893834, 1], [2.209717944343731, 2.566803071772211, 1], [2.208364779650318, 2.564397445650587, 1], [2.208364779650318, 2.564397445650587, 1], [2.208364779650318, 2.564397445650587, 1], [2.208364779650318, 2.566803071772211, 1], [2.209717944343731, 2.566803071772211, 1], [2.211071109037144, 2.569208697893834, 1], [2.209717944343731, 2.569208697893834, 1], [2.209717944343731, 2.571614324015457, 1], [2.209717944343731, 2.571614324015457, 1], [2.209717944343731, 2.571614324015457, 1], [2.209717944343731, 2.571614324015457, 1]],), ([[0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.606217782649107, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.604864617955694, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.60351145326228, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1], [0.60351145326228, 2.564397445650587, 1], [0.602158288568867, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1], [0.602158288568867, 2.569208697893834, 1]],), ([[2.414045813049122, 3.343820309056582, 1], [2.408633154275469, 3.310141543353854, 1], [2.408633154275469, 3.305330291110607, 1], [2.411339483662296, 3.290896534380866, 1], [2.411339483662296, 3.269245899286255, 1], [2.414045813049122, 3.254812142556514, 1], [2.414045813049122, 3.247595264191644, 1], [2.412692648355708, 3.23797275970515, 1], [2.414045813049122, 3.213916498488916, 1], [2.414045813049122, 3.194671489515929, 1], [2.411339483662296, 3.180237732786188, 1], [2.40457366019523, 3.158587097691577, 1], [2.40457366019523, 3.15377584544833, 1], [2.405926824888643, 3.14655896708346, 1], [2.40051416611499, 3.129719584232096, 1], [2.397807836728164, 3.120097079745602, 1], [2.397807836728164, 3.122502705867225, 1], [2.395101507341337, 3.117691453623979, 1], [2.392395177954511, 3.112880201380732, 1], [2.389688848567685, 3.108068949137485, 1], [2.389688848567685, 3.100852070772614, 1], [2.389688848567685, 3.096040818529367, 1], [2.392395177954511, 3.08401268792125, 1], [2.392395177954511, 3.064767678948263, 1], [2.392395177954511, 3.045522669975275, 1], [2.397807836728164, 3.028683287123911, 1], [2.405926824888643, 3.023872034880664, 1], [2.414045813049122, 3.0070326520293, 1], [2.420811636516188, 2.99259889529956, 1], [2.428930624676667, 2.980570764691442, 1], [2.438402777530559, 2.963731381840078, 1], [2.446521765691038, 2.94448637286709, 1], [2.457347083238344, 2.927646990015726, 1], [2.470878730172476, 2.910807607164362, 1], [2.476291388946128, 2.896373850434622, 1], [2.487116706493434, 2.877128841461634, 1], [2.497942024040739, 2.86028945861027, 1], [2.514180000361698, 2.84585570188053, 1], [2.525005317909003, 2.829016319029166, 1], [2.538536964843135, 2.814582562299425, 1], [2.550715447083854, 2.795337553326438, 1], [2.562893929324572, 2.780903796596697, 1], [2.573719246871878, 2.766470039866956, 1], [2.589957223192836, 2.752036283137215, 1], [2.604842034820381, 2.737602526407475, 1], [2.6170205170611, 2.718357517434487, 1], [2.633258493382058, 2.708735012947994, 1], [2.648143305009603, 2.689490003975006, 1], [2.663028116637148, 2.679867499488512, 1], [2.668440775410801, 2.670244995002019, 1], [2.687385081118586, 2.650999986029031, 1], [2.699563563359304, 2.646188733785785, 1], [2.709035716213196, 2.63656622929929, 1], [2.721214198453915, 2.631754977056044, 1], [2.73609901008146, 2.617321220326302, 1], [2.746924327628766, 2.612509968083056, 1], [2.756396480482658, 2.600481837474939, 1], [2.767221798029963, 2.588453706866821, 1], [2.773987621497029, 2.581236828501951, 1], [2.780753444964095, 2.574019950137081, 1], [2.788872433124574, 2.571614324015457, 1], [2.801050915365293, 2.55958619340734, 1], [2.809169903525772, 2.549963688920847, 1], [2.817288891686251, 2.542746810555976, 1], [2.828114209233556, 2.535529932191106, 1], [2.836233197394036, 2.535529932191106, 1], [2.841645856167688, 2.535529932191106, 1], [2.852471173714994, 2.521096175461365, 1], [2.857883832488647, 2.516284923218119, 1], [2.866002820649126, 2.506662418731625, 1], [2.872768644116191, 2.501851166488378, 1], [2.88494712635691, 2.492228662001884, 1], [2.894419279210803, 2.487417409758637, 1], [2.898478773291042, 2.48260615751539, 1], [2.912010420225174, 2.477794905272143, 1], [2.920129408385654, 2.472983653028896, 1], [2.925542067159306, 2.46817240078565, 1], [2.933661055319785, 2.458549896299156, 1], [2.941780043480264, 2.456144270177532, 1], [2.95260536102757, 2.448927391812662, 1], [2.958018019801222, 2.441710513447791, 1], [2.966137007961701, 2.439304887326168, 1], [2.971549666735354, 2.436899261204545, 1], [2.976962325509007, 2.432088008961298, 1], [2.985081313669486, 2.424871130596427, 1], [2.993200301829965, 2.420059878353181, 1], [2.998612960603618, 2.41284299998831, 1], [3.006731948764097, 2.40562612162344, 1], [3.01214460753775, 2.39840924325857, 1], [3.014850936924576, 2.396003617136946, 1], [3.022969925085055, 2.388786738772076, 1], [3.031088913245534, 2.386381112650453, 1], [3.036501572019187, 2.376758608163959, 1], [3.04191423079284, 2.376758608163959, 1], [3.047326889566492, 2.374352982042336, 1], [3.052739548340146, 2.367136103677465, 1], [3.060858536500624, 2.364730477555842, 1], [3.066271195274277, 2.359919225312595, 1], [3.074390183434756, 2.355107973069348, 1], [3.077096512821583, 2.352702346947725, 1], [3.079802842208409, 2.347891094704478, 1], [3.082509171595235, 2.345485468582854, 1], [3.090628159755715, 2.345485468582854, 1], [3.096040818529367, 2.338268590217984, 1], [3.10145347730302, 2.333457337974737, 1], [3.104159806689847, 2.32864608573149, 1], [3.109572465463499, 2.326240459609867, 1], [3.114985124237152, 2.319023581244997, 1], [3.117691453623978, 2.311806702880126, 1], [3.123104112397631, 2.309401076758503, 1], [3.128516771171284, 2.30699545063688, 1], [3.133929429944936, 2.302184198393633, 1], [3.139342088718589, 2.302184198393633, 1]],), ([[1.764526760210793, 3.271651525407878, 1], [1.73746346634253, 3.247595264191644, 1], [1.661686243511391, 3.199482741759176, 1], [1.661686243511391, 3.199482741759176, 1], [1.615678643935343, 3.177832106664565, 1], [1.588615350067079, 3.165803976056448, 1], [1.577790032519774, 3.151370219326707, 1], [1.552079903344923, 3.141747714840213, 1], [1.526369774170073, 3.136936462596966, 1], [1.511484962542528, 3.132125210353719, 1], [1.497953315608396, 3.129719584232096, 1], [1.477655845207198, 3.122502705867225, 1], [1.458711539499413, 3.105663323015861, 1], [1.435707739711389, 3.100852070772614, 1], [1.415410269310192, 3.096040818529367, 1], [1.39781912829582, 3.088823940164497, 1], [1.374815328507796, 3.07679580955638, 1], [1.359930516880251, 3.050333922218522, 1], [1.350458364026358, 3.026277661002288, 1], [1.340986211172466, 3.016655156515794, 1], [1.324748234851508, 2.99259889529956, 1], [1.30851025853055, 2.966137007961702, 1], [1.294978611596418, 2.954108877353585, 1], [1.282800129355699, 2.93005261613735, 1], [1.267915317728154, 2.910807607164362, 1], [1.257090000180849, 2.891562598191375, 1], [1.250324176713783, 2.877128841461634, 1], [1.236792529779651, 2.8530725802454, 1], [1.223260882845519, 2.83863882351566, 1], [1.2110824006048, 2.821799440664296, 1], [1.198903918364082, 2.809771310056178, 1], [1.18943176551019, 2.790526301083191, 1], [1.178606447962884, 2.77609254435345, 1], [1.165074801028752, 2.766470039866956, 1], [1.158308977561686, 2.752036283137215, 1], [1.150189989401207, 2.737602526407475, 1], [1.142071001240728, 2.723168769677734, 1], [1.136658342467075, 2.708735012947994, 1], [1.128539354306596, 2.6991125084615, 1], [1.120420366146117, 2.684678751731759, 1], [1.109595048598812, 2.667839368880395, 1], [1.102829225131746, 2.655811238272278, 1], [1.09606340166468, 2.641377481542537, 1], [1.089297578197614, 2.631754977056044, 1], [1.082531754730548, 2.626943724812797, 1], [1.074412766570069, 2.614915594204679, 1], [1.064940613716176, 2.598076211353315, 1], [1.063587449022764, 2.583642454623575, 1], [1.051408966782045, 2.583642454623575, 1], [1.047349472701805, 2.574019950137081, 1], [1.040583649234739, 2.564397445650587, 1], [1.0365241551545, 2.55236931504247, 1], [1.028405166994021, 2.540341184434353, 1], [1.021639343526955, 2.533124306069483, 1], [1.017579849446715, 2.523501801582989, 1], [1.009460861286236, 2.521096175461365, 1], [1.005401367205996, 2.511473670974872, 1], [0.99863554373893, 2.501851166488378, 1]],)]\n" + ] + } + ], + "source": [ + "trajectory = filtered_world.get_traj()\n", + "print(\"trajectories are\", trajectory)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "afc827d0", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "get_traj SELECT asMFJSON(trajCentroids)::json->'coordinates' FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM item_general_trajectory WHERE worldId='a29b573e-70a0-4b6c-b19d-4f453867ce4f') sq0 WHERE sq0.objecttype='car') sq1 WHERE overlap(sq1.largestBbox,'STBOX Z((0.01082532, 2.59647246, 0),(3.01034039, 3.35985782, 2))')) as final\n", + "headings are [[None, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 240.64224645719918, 0.0, 0.0, 0.0, 270.0, 105.70863782901449, 254.29136217098548, 0.0, 0.0, 0.0, 0.0, 90.0, 240.64224645720623, 74.29136217098551, 270.0, 0.0, 90.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 90.0, 0.0, 0.0, 270.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 299.35775354278474, 0.0, 318.36646066342803, 119.35775354278525, 299.35775354278525, 0.0, 270.0, 0.0, 270.0, 0.0, 90.0, 270.0, 0.0, 0.0, 0.0, 90.0, 119.35775354279377, 119.35775354279377, 270.0, 90.0, 270.0, 0.0, 0.0, 270.0, 0.0, 0.0, 90.0, 0.0, 0.0, 0.0, 285.7086378290145, 90.0, 0.0, 90.0, 270.0, 119.35775354278576, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 90.0, 299.35775354278474, 240.64224645720623, 299.3577535427938, 0.0, 0.0, 90.0, 270.0, 0.0, 0.0, 90.0, 0.0, 270.0, 240.64224645721524, 0.0, 0.0, 90.0, 0.0, 60.64224645720623, 0.0, 90.0, 0.0, 0.0, 0.0], [None, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 285.7086378290133, 105.7086378290133, 0.0, 0.0, 0.0, 0.0], [None, 260.8698235177206, 270.0, 280.61965527615735, 270.0, 280.61965527615394, 270.0, 261.99527114270296, 273.2194948968547, 270.0, 259.380344723846, 252.64597536373927, 270.0, 280.6196552761549, 252.18111108547583, 254.29136217098616, 90.0, 240.64224645719815, 240.64224645721475, 240.642246457213, 270.0, 270.0, 282.68038349181927, 270.0, 270.0, 287.81888891452274, 329.3493320429478, 295.74070835623223, 295.1148348861449, 304.01934998982523, 299.3577535427904, 292.8736651906257, 302.73522627210974, 308.7843641002978, 290.55604521958094, 299.3577535427917, 302.7352262721065, 318.36646066343224, 302.7352262721065, 313.152389734005, 302.3261399943782, 310.1559996249172, 306.8698976458454, 318.36646066342814, 315.88140399658323, 302.32613999437694, 329.34933204294947, 307.7197109160052, 327.1187603347254, 299.3577535427924, 314.5488614532131, 338.442920714965, 314.54886145320773, 338.4429207149664, 315.8814039965797, 336.0375110254266, 308.2204320548504, 311.98721249581337, 313.1523897340067, 313.1523897340068, 343.4956386182417, 315.3558698306227, 310.15599962492115, 318.36646066342706, 326.30993247402074, 0.0, 0.0, 306.8698976458443, 318.3664606634374, 310.15599962491723, 324.58294472352856, 321.68690932618284, 333.07232214896084, 310.155999624917, 340.4268741695884, 329.3493320429492, 318.3664606634327, 310.15599962491723, 343.4956386182417, 326.309932474024, 306.8698976458361, 343.4956386182474, 336.0375110254275, 318.3664606634321, 318.3664606634255, 329.34933204295106, 306.8698976458429, 318.36646066343076, 306.86989764584797, 318.36646066341694, 318.36646066343076, 343.4956386182474, 299.3577535427921, 0.0, 336.03751102542225, 306.8698976458491, 343.4956386182457, 318.3664606634295, 329.34933204294646, 318.36646066344207, 299.3577535427893, 318.3664606634123, 0.0, 306.8698976458412, 318.3664606634321, 299.357753542795, 336.03751102542225, 306.86989764584797, 290.5560452195796, 336.0375110254257, 336.0375110254275, 318.36646066342513, 0.0], [None, 221.63353933657015, 212.41230661835067, 0.0, 205.20112364547506, 203.96248897457784, 233.13010235415794, 200.5193171608242, 190.5994237164336, 197.91250938596315, 190.08059798754033, 199.57312583041212, 221.63353933656953, 191.81314470013723, 193.33510634104744, 202.3062050549055, 207.60397993061923, 240.64224645720856, 248.50795996665428, 225.45113854678763, 235.98065001017392, 238.46520809481171, 221.63353933656973, 243.14918670604567, 232.28028908399415, 240.64224645720915, 244.88516511385575, 240.64224645720833, 226.84761026599236, 234.1250086479352, 224.64413016938153, 243.79416530447276, 233.1301023541551, 215.41705527646823, 244.88516511385575, 240.64224645720864, 240.6422464572095, 249.4439547804146, 229.84400037508198, 240.6422464572095, 237.26477372789347, 240.64224645720833, 244.88516511385575, 234.88798881557486, 215.41705527646698, 235.9806500101755, 240.6422464572073, 264.6441749571486, 0.0, 247.12633480937143, 234.88798881557736, 251.35046124595095, 235.98065001017426, 226.84761026599236, 247.12633480937143, 196.5043613817579, 247.1263348093697, 234.88798881557858]]\n" + ] + } + ], + "source": [ + "headings = filtered_world.get_headings()\n", + "print(\"headings are\", headings)" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "aca1cfc3", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "get backend Qt5Agg\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAADfCAYAAAAN+JPJAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOz9WYxlW3rfif3WsKczxJiRkeO9t6puFWsii8Vm05So7pbdNtBu2BBgoAW1DaMNCCAEyBAM+oGUX/pJgJ4EGLBfaLhhNeC2JHhAN4zulluEZolki8VisaZbd74354z5DHtagx/W3id2nDjnxInMvFSKyO8ib0TsYe21117rv77v/33rW8J7zxt5I2/kjbyRP10i/3VX4I28kTfyRt7Iq5c34P5G3sgbeSN/CuUNuL+RN/JG3sifQnkD7m/kjbyRN/KnUN6A+xt5I2/kjfwplDfg/kbeyBt5I38K5QsDdyHEfyCEeE8I8YEQ4re+qOe8kTfyRt7IG7ks4ouIcxdCKOBnwP8EeAD898B/7L3/8St/2Bt5I2/kjbyRS/JFae6/Anzgvf/Ie18Bfwf4C1/Qs97IG3kjb+SNzIn+gsq9C3ze+fsB8D9YdnG/3/c7OzsLzwkhXm3NlpTfWjBf9PNeRF64Tmved+Gqq+7xnkW2nrjG866qx9Ly56vyAmUvK0MsOLayvOYiL1aXv0oWWs3/uvrfXF38kuPdvviva3X7v0mr6l+2rlfd/+DBgwPv/d6ic18UuF85FoUQvw78OsD29ja/8Ru/0T238Pe5+5dec1WDCCEu/FtW7qrnX+e8lJcNJOfcWuV672dlrFsXN/f3peua9vHN+bbsdSY67/3Cuq9Tv0V1vVBGW/78tWuXevEez3mnUwu+NYBt3lV13t0tmcC6dZkH96XiL9636GUu9Nduv54rZjYRzX2n+XIWvacQYmn7LirDOYf3/sIzBCDnwL1tP7nguS8KbN3nLnvXrrT9sXtNO24ulIsHD6Jt2blmepH6ts9p750vo23HZc9Y9sz5cpfd8xu/8RufLqvbFwXuD4D7nb/vAY+6F3jvfxv4bYD79+/7V6kxX1XWIrBddN/L1qkLmPNltXVYBJTz9813kG4ZHnDtQFhR9+5gcc21wnsQAu8cLGiT+QEyXwcp5ZWT1DJZNPjmZZUm355fCIANOHfbcNE3n9WhO/g6Za871K+67sqyhFj4HvP3vEh/XNinrwCxV6UZv4yWPw9sy97dez/rh917VraV55UA+5XP4fJ7rLp/0fsumtzW7QdfFLj/98BXhRBfAh4Cfwn4X35Bz7ryg84P7GWazarzi+RlwL+9dxFALpsIriPe+4VUyUzDoOnf3uMbgJ9vg0Uds51o2jpJKZdqLVe1z1Ud9ipwX1puW14DmleCTDPhLXv+Qlnz0/vmv3DL/HT0auUq5aRr0S3TjNs+s3BMzWnys+c097wKab/dOpM/nWsXAfzaz/yCaZ759l72vFVWyjoWzLx8IeDuvTdCiP8t8PcBBfxn3vsfrXPvq9Tgu6B4FaB/UTIPmIsAsAXIVR16GUUCjVd8Tvu7dH/3vmXnugO80erb6+fvaes6o4LmJqj2nWb1bq+/+FIX6vNiNsBlWdVW6zxjVTvOK32L4HqBLUZDClzJ/S8TNXsngRfLrSrZ/SZraozdWtJo9l3LctZnm/7bvoe4UK/rSbeO3Vq+EAXXAfh5RePi5ARefLFAvqhu7c92jKyrzS8D+XUB/ovS3PHe/9fAf73OtYu471XXriPX5YCvU/5167qKnukO0mXXdMtb1GlXm6BLNK0Vf3frBsvBvdvJ5i0Q7xyiO7kuee6l2nVog+7EsEjmJ4tXoYHNa64Lzy+ox4U6zPHUYRJbW9Vf+rALbzuvcXf7GlygpeaL6z6ie3/XkhMspq0WyTpa5XU16lU0zKryr6b6xBdpPC1V4LrWLlz0Eyxrm2Xtum5bfmHg/rLyMlp1F9jX4cS+CFk2ucwDwTqd8kUoo1XyQpPeooMdbXyRVnoJ0ub644yuWKLRd8u5gHl/whZXK4u0/xYQL/D713nQkjHa1fN9KPhi3eb8Cpfq3znnGoruKppr4WTbOUdbjyV02nX45fljyyi6hQ5FLs+B3fddee+adM/LyDKA79ahC/TdOi2iOFdNGKvktQX368oX+cHWdtQsufeqcy9T9xdxVF1Xlt0hCBo6QgRtlzlA77bZqnZYUTc/d83Ker4A37qgkJWWwqVohwV1W7eNvfdLX0zQoVe4CF4t2Lfa9fyzL1gz8/c1Za3TTldplNe5Z51nrTo+Tw2tU1aXCln1nFUT+bqySGmbL2sdyrVrEa8D8KvktQT3FwW7VfTOdemWVWFV64DyIipmXc1lXbkuR9+VVnPodqSXqUvL77Za1VVExCXNawWwyyX16t7RarNyrpx5sLuqF5zTKJ16XPHcF56aFw3YzvPX+QoX+mXz8wLtsuK+mdN9hUghsF3lZs06rduHZt++y+UvAca2vvPhmIvq1KVC1uW65++fr8O6Mv+cS/TYiv7VHYOLyvnX7lB9UXlZ7XsVuL9sXVowvIoiedEOtKwDdD/2SmfRkvosknnaquscnZ/U1ql71ynWHpdXgJ5Y5V1sr1nxzGV1WjbQl5XZlav43GXPeGFtldV96Kr1CpfK4uLkul4lxGzdw6IyoeMch6URM8uLX0Nh6DiD1ylrNoktUW7mJ4T5gIWXtXbXtXqWyVWUDSwPM75O/V8bcF9r4DYvtmg2m5/x1nneqtl1/vgicF9WhxcxoRZd275zW5619tplwWWn7TKLYlHbXlXHrlYEF6M11tLyllZ69r/2gWuUdrFur1Kuq42tI6soDyEEihXau18QZTJfp/bS5ueM5W3adh46ltVHdH7i/dJ1F4veo/v3Mg58/vy80nPdyJJVZc2HS74IB/+ik/l8nVtZ1q+WWQ/r1ve1AfdVctXsf12NfR1apZV5YFx233UH+zrn50M5lVIrtZV5oF31zK4mM9++63S+VZPeWl+ivWhZO4i1S/oTky9i0ljnWdeBkWU+ilarv+AM5XqUg/cevwawX0eW9eWrLORF16yiMLoK2LqBDOvW+zqW+qoJfZVCuEixvUpeW3Cf1zCXvdgi8O2Wsei6F6nLdY63z1tWz3WeteidVk1yV7179/dVHWQRWHcnglV1elHtdZmmtez6q667rlb5Rcqq+i6zBpZNlBc09WV9snOt6PzrlnGVk1rAJavAd39eoWwtkpd1UC46vwwM59t8vo1b7X1Rv+s+d93V19ex1Fdp/YsmnkXjsPv3KnmtwX3RytJlq9DWoRBetB7XtQy6z1xmHq66Z5WsA8xX1aXlIeeft+r+l7VEvqh7XsREXjQRvUpttCuvqt26gCw6vy8qb9m7rHKKXgDA+Wu8D//mrLwuQF6lWV6nvbtjf9Wq06s03nnKp/3ZVWBWgft8fdf1QV0l60yOiya3+XF41eTz2oH7dTWyZcCz7rGr6qKUWgluV3Fh65pv8wOle8+r1jRfJJ3Bq5JX8V1e9r55+aKB/VVMPhfOt+UvvvlCOcvKX/X3ouvnabx5ZaXtu61faF6uok3my2qlu+p0WV3XAf9F2vBVWLNs0dGy91sl16Fu5u950X702oE7XP7A3c7Vvab7+xcBgq8aZFfdv8phu2zmnj9/nee/CkC9Tmd7laD+sve+zHPWde698r4YCr4Yqz5vxTTPXjda5kWoklVUwary1qESF51fl2dep18u0tQX1WORgnVVkrzW2ug++0WT6i2qb7ee61KYrx24LzOZlzkLF83CXzSILZtJrzuo5zWhqz7aq6ATXqWWvG5dln2zV1WPly3jRUzqFy3/RayELo9+QWtvgNxznl+IzrVLOfU5muLK5y8A9HngelXWz6IJ5EXK6NZpXVC/6v5VFu91LIqrnr+OhbPq/lZeO3CHy062eWfevIb0RQF724kXde5l972I1nadOi+bAK7jg7iqDvPlrVvWOprYsuPX1c6uI1+klr+uFtWty8sA4TJ46YK5nHvGi3zLrizT1P+kpNWaX+a5L/re6yhc3fPzuPUq5CoNfpm8tuA+//sirmyVhvSqtMUuWK8L3OsO+HktZdnkcJXGvopDfdWAuG7HWlXWKivgRSapLwK8FznjXtXz50Fjnba8anCLuXPXqct8Wdepy6sob/6++XH8oiB5Xc35unz+PDa8KkpmkSL5IhPsawnuXVnGNy8DiJflO+fvnV+mv+zZ7TWLTN7rTAjz5bR/d2mpRZ2yvba7Gm+RGbmss75Iu131futQOdc9/yclbTtNJhMODw+JooitrS2SJLkylfS65b+MzNqey/TLKjB6Ff6A+Y1mFilezrkXArhXWc9ufa7z7JcpYxnwrkPBvOo6vbbgvqxjXAUkr4KWueqZqzrevEl2nQmnfecWoC+k0F1i6s2Xe9WAuq7Ta/456/CB1wH7+We/zDe7vtbftmnnnuZoXVUcHR8zOjujyAuKsuDx48dkWcbOzg4bGxskSfJKLMT5tl3296zWV7T5MjP+ZSmh+efMT3KLzs37ypbVvauULLvuT0JexJLq3tsdf4v6xrJEZleNvUXj7qq+9tqCeyvth39Rc/M61y7Siq/S0teVVeDVBfXu3y/aydrnvArA7D7nOpPcfF3a6xcdX3TNi1oR1xfX+ekxxnJ6csbJyTG1MSEHjrMIB6auGdU1o7MRWiuGGxtsbW0xHA6J4/iClfcyIL9K1qXcXlYLfZG+Pf93dzwt3Pzjiue8DmC/TJZZGKt2fZu3prvvty6FdJ3x8dqDu5Ry7bjsFx1Qi0B9nfLW5WJXfbjux3rZ0KllZXfr8qKyzqR33TZs208pBVye1NbpyN13tNbivZ+tTVhfwrL60XjE8eERRVF0T+Fdk9On80xTG46Pjzk9OUVpxXA4ZHd3l+FwOHufRQP3RZSObhnXsX5WafzLtPtlfy961nydVmnlV4HXPJ3xKsH8OhbLq7Rs1jn3Ir6Jde+5EtyFEP8Z8D8Dnnnvv90c2wH+LvAO8AnwF733x825vw78ZcACf817//evU+n5WXBZR72uLKMV2p+visJZVPY68iq17Hlt4bra2jr3rDNxrKvVd/+1dNR1Or21loODgwuAk2UZWZYRx/Hy+vjwHvl0yuHRAdPpBKyfbR1nrGU8GVPkBdYLdBShtUZKgRAS78E5i3WWw8NDjo+PieOYjY0Ndnd3ybIMrV+d/rSOBXSd+xcBxSprav66db5Rq5xdde0i7f9lLONV4+8qRes68qopr1c5qa3T8/5vwP8J+M87x34L+B3v/d8UQvxW8/dvCiG+SdgM+1vAHeAfCCG+5r1fvGytkS7nNg+07bku0K/zcebNphfRzK973auQ64L8VVrhi1g9y36/rlz1Lsu+SZd/XcfhNB6PGY1GbG5uzu7Jpzmj0QitNb1ej42NjUt1q6qKk6MDzs7O8N5C01ecs+R5QVVV9HphgiiKmsrUYQIA4jghjtPQ/pwDUVVVHBwccHR0RBLHbG5tsbW1Ra/Xu/Qt1nm3V3F80XXr0gGLyl2XLuyOu/Y5i8IaV/kW1rFClt3TLXdVOOVVx16kXVcdWyavmnq6Ety99/9ECPHO3OG/APz55ve/Dfwj4Deb43/He18CHwshPgB+BfiXVz1nEa8+DzLraMPdxpzntBaVOy8vqhVddc8XPUGseuZVz34R6+i6ALGsnsuoh+to8VEUEUVR8IR6QRsNLqWalYEXICTgsNZycnLCyfExti7BO4QAvGc6HVOWFVnaY2Nji8OjA3763nsMNze5dfs2vX4P7zx1bZhMJggpSOIEpAQ8QgrwYJ1lanLyac6zJ09Js4zNrU329vZIkuTa7bTutS+jfa4j1/nOixQ14JLzdH7MztdnfuPrdSeX7jO7/WhdgO+e+6IUwatWvr6MvKjNuO+9fwzgvX8shLjZHL8L/G7nugfNsWvJMlNqFQgtoxSu0mznj68DVMvK+yKonWV1WTTxXff5L6qhL/oOy+p61TOWTabtd1xHi/fOI2W41lp/Hv7iBYg234nAO8d4dMrR8SFVWYUFai44UvNiSlEUJEnC9tYWx8dn/O7v/h4//ulPmEwnCCkZDDe4ffs2b99/i/39fbKsBwiMMZiyxHtPHMdorc8tR8AYw3QyCZOBENy9e3dpe6wj8338RbTFF6EArkvPrFvH7mS+SlNfVd9lbbkI0Lt9bB0Loj13FXOw6vnLjl8neKKVdb/dq3aoLmrhhbUQQvw68OsAOzs7Kwtd9WGXAew6ptYi4Onytovu7XbEZXV4lbJO2avy0rxMude557oc6aJ7F32D+cHexk9fAA48WkrK6RRjHHEcI4QI3LkUCDzTySnHx0fk0+n5twSquiSfTpCRZnN7i9HZmH/1z/8FP/rRD5mMJ7NNvJ2znJ2dcHZ2yvs/+xn9wYBb+7d4++23uXXrFmma4gn8/7QoEYDWMVqdAz0CcOsB8Xw/vi4Qv0qAXzX2lo2Xdazr+VDKeaDtAuCLUhbLxvFV1/1JyCJgX3ccrXPNi4L7UyHE7UZrvw08a44/AO53rrsHPFpUgPf+t4HfBnj77bd9c2yldr1IK19S9qV7ls3ci0B+kWa0bBL5IoB9HRplXst4VR1i3ed3f76MWdnSJldZZavih7015GfHjM9OyYuSSGvSLKPfH6LTlMo5TqryAkhYaxtaBYYbQyb5lN/9vd/jj3/wQ85Oz5ACelmCcY6yqs77j/c47xmNThmNRnz40Yf0ej1u3rzJO++8w61btxgMBuDB1JY8zxFCoLUmjuMAWq6hbxqZ9yctaotutsV1Lct1ZBXAfxHaZPf6q47Nf/eXkXUmui9CXqRdXkZR6sqLgvt/BfwnwN9sfv6XneP/hRDibxEcql8Ffn+dApeZPV0AWXdl4Dz4XbexlgH7fL3WtSbWfVb3me25RR1+ZvIvmJyuW5/rUktd4FknCmId8T5QJ1rrhZpqV4u73Dc8k/EJxfQE72qwFbWtqMspk7NTNnf3UFojZKDcnXVMJhO89wyGfYqy5Hvf/yO+/0ff5/j4BIVgazjg61//Ob767leYTiY8e/6cBw8ecnB4yCTPsQ6CveBw1jEenzEan82A/tbNm7zz9jvcunOXfq+PQFLXNZPJhE8++YS8yPnq1752gU6bB3TRpASz1mKMwfvgxBVCYJvQzFelba6iT9a5dxmdsUyrv06d5leIv2p+et36veh7tPfCcmpq/vw6+PFKaBkhxP+D4Dy9IYR4APynBFD/e0KIvwx8BvxHzQN/JIT4e8CPAQP8VX9FpMyyCi8y0a/z8os63VWmYxc453+u0t5fpSwc6Eu0nFeprbdylfU0TwGtW491nrsKDBYdD9eD9xaPRQiLkE0/8YDwM57QOst0OsVaS5pmeOf40Y9+xB987w9DGKWA4XDIt37uXb75tXcZDAZ474mVYHMw4N133qEoCp4fHvLw8VOePn3G6fgM6xwOj0DgkEwnEz7++CM++eQT0jRj7+ZN3rr3Nnfv3mM4HAKgtJ7FwnffpSt1VVAUBXmeE0URpycjkjRld+8GURSt3MTiZeRFgH7Z+Jo/tq72vAgHulTOvGLzRTkku/Iymv86tFt77DpW1FV1Wida5j9ecurfX3L93wD+xlXlXvFM4HKDdc33ZXlTVsm6puC8mdxqll+0vAqgfFmr4br3rqudXTUxrwrZXDQQrrLGPOG7GWOYnI0w3tDv99Ba88EHH/CH3/s+z549x1nD1nDAu197l2998+tsDDOkcAgE1gqkVCAF1hmyfspbvbvcv3uXqq45Oj7h8dMnPHr8iJOzEZV1Yb2rDzHweZ7z2Wef89lnn5MmKXt7N/nSO1/irbffnr3DfLs55zg9PcFU5QXFQilNXRvGozFxHJOkydphri8qq8C2Ky9D3yyjRhc9o3tdt4z5ld3Lnjt/38u8y+V72789527G9l1bH4/A+/a7N/e482v8hXvb+rRltpNct9+sHr+v/QrVeWk52nYWX/TBV2m8y2QRqM+ff5Va0jKrYZXWvAzQrmPRLHv+Klnk61jGDc+XPz9JrvOseUtplebT/DX72V6rlWY0mvAvfvcPePDwIds3tvnqV7/K7u4Ojx49xGPZ2Ojx7jtv8e1vfZPBcBAGn5CMxmOKomB3ZwcDPHz8GK01N/f2kICQlkhG3Lp1k1u3bvIL3/o6p2cjnjw/4NGjJxwcjZgWxYXBKiRkWcqNGzfwhIRkaZpe6r/GmIaGuaiZllWOMYazsxOKoqQ36LO1tbU0x82rlHWUpnUm+RcZl/PPWFSnRVr9OnXoRjUB1HUNrLY22vu7E0lLneE9SmtCOG4Q1wC5EM29Hpy3oa819B7egQBBO0lZ6IB4i0tdn5Ex5t+8bfaWmR/z5uL8i79sx1mWv2ZVZ1kEtlcD0UUgX0QdLdJiVr3XKnC9iorqHl/2nldpVfPvAlz4Lu0559xSC6g7YS9yngaqxTdcdGPFSUkYBQ28C4FAURQlP33vR/z4xz9jnOckacpb/R43drZ58OABn3z0KXdu3+Q7/86vcePGdhhoaE7PRhyfnvLez37G3vYON7Z38N5zcHSEdZ6nz044PjrhG19/lySO2djoI4VA49jZ3mJne4tvfPUrjKcFT58/59Hjp4ynBfffeod3vvxlNoabSCF59PAhz54+5e133mF/f/980Lsw6EXTjHVVUZZhMZW1AXicD6GXo9Mzzk5PUVozHAzZ2g5A312R25Vu/143+mTV5HrVPa1ch4pZpsAsAttF/XUppTRfBQFKqUt7qc5TPO3YX9Re1oZFb1IqnLPUdY0xNSCI4whrLdbWSCnJ8wJjDFrHSKGw3lBXNXVdhdXPtkYIQRQleB8mmbIsm3QabuaLGo1GTfkxR0dHV7IJrx24L5XWDBHdQ4Eu6QLgOmA/f24Z4HUnkUUTzLo0z/w93fpdtan1VceuklVg/iLlz2vUV00K89/mOlkrLw9kh7OW0WjEs2fPEFIz3NigqksQIJF89NGn/MH3/pDTsxFxHPOtb3+Nd9/9CofPj/js00958vQ5N3a2+dbXv8budgDbUV5ydHzMxx99QlHkfPObX2ezv4FAIIUIi5d6PZ4/PcVZzUeffsrz58/5+s99lW/+3DcwrrqQT31j0GNj8DbvvvMWVmh0MkDI6PxbezC1wVkbQEZI8A7jPaaumU6mVMUEa02nDVx7K+Bn5n1dVhyVhxwfHxHFMZubm2xubjIcDsPCrk7btxLWA1xNM7Zjqc3Zs0qW9YPr9qlV/amraK17zazcznkQDfieO6attRdCE51zVJ0oKWMMVVVRFAWj0QhnbUPdhair6XQ6u14IQVEUlOWUOI4Zj8c452apKIwxs2e3VLNS7YK7FiMUUp5r/O17aR2TJBctvmXyeoF7872EnKv4jMK6yDPNf+CuNr8sWVHXhFukcS4qs/v7Op2wPd/9OX/vdTXxVbKKqnmV0nb6rjNwHoCXTnrNNZHWGGupyhKl9UL/xrJnTycTDg8OKIqc6TSnrGqOj4+JRI0mmMB7ezv88i99h9HZiDv37jKa5vzTf/q7TCZT/uyf/VX+x//DP0+SSpSQSEAJwaPHTzgbTbh37y7eCXZ3t1HIttJEOiHSKUlSc+PGFlFyg/v37uC8RSoxSxN86Z2FJIkjhJTU1uAsICRSBq3x5PiErNdj0O8zHY85Pj5qgMAiG0pHiNCexgYAqo2jKmtME13UAob3UJUlz58/4+D5c5I0ZXNzc0bdzOdpuqr/voi8iKbf3neVxb5sjC5SMLrg7n1ICmdq0wBueYF+qeqa8WhEWYZQ2TzPQ+I4ITg9O8V7TxRF1HVNVVVAWBEthSSOIpIkIY7jWWbQ9u8woYJunOfzlmw74V483rahRAjZ/N0NJJHAxUSK/+3/b3nqrtcG3F+EMVwGavNm/SJAnl8Is6iDreL4lj1/Wd2W0S5XvU8XNNd99vz5+ffwPuRVMcYQRdEsi+Iimmi+zFV1n6+bc47j42O898Q6Io5jnHM8ffqUg8NDdKQZDof0+336/T5xHC/M6FiWJQcHB4zPTnDONEfdzHclRKBrhIDhxoDhsI+1lsOjIzY3BvzSL36Lm3s3uHv3LkIojJ0iZ11E8Pb9t7HOcHpcMBoXVHVFP8sCtnuQUjcD6zylQb/fx1iDFAGGL1MJ4BHY2qCoiKQGpXBe4WxISvb82QGHh4dsDAco0eHnvQcZ+rExNc55pNQoqbEYcJYk0lhnKYpqpvlprcEHda/IQ7TN8+fPSdOUL3/5yxfy66xrcb2IXHXvKg1/UV+dv65rTcO5Vt1q25PJhKIoePr06YwKnI7DMSHO1xy0k2M7BqSUbG1tkaYpw83gy9CNAtL9p5RCAEqqtd93fvJZvAetw3s3A/E2DfX5ZAXdjRb/zdHcxTnlcqnS1+xnbeMt2kWpLX9ey1wG5POyKAXxsnsXTRar8j0ve5f5v6818DyIi1wWdVXx+YMHHB4eorUmyzL6vT69fo8sy0jiBKUXp82d900s4iWllDjvefjgAZPJhN3dXaZ1zcHB84Y/thhTU1Ylk0lIxKWUIo5j+r0+g+GAQb9PEsecnZ1xcnIanEzOgfPUdU1dVHgXHKDn79iERgJSSfb29vAIdna36WUZnkabQiOkxtkaITxpklKZCq09cSQZjwr6aY9WeZe0nKxGCY2xAfClcyztnE13boFfCEdVlgTNK0aJEHvvCAAuIwU+LJByzuGNxeOJdERdG549fcaHH33Ep598RlmV3L1zl3v377G3v08cx1hrm1TFAq1jlFYzB+N0OqUsy1CtrrIw61pNn+polN3v+8qktcznbJ32t0V93drAZ4eJzjAejzk6OiLPc5xz1HXNeDyeKSotGGdZxubmJv1+n16WkaYhAVwUBUBvVw7Tjn/ncL7x6whBFEcgxOy5l6zKpv1EB7fat5u9VedyIVpwDiKlwM1WK7chvYujgTo96VL7rJLXB9z54nnmrqwCyWXnlqUgXkcTXxeU1zVt1wZ5wcxBB1CWFY8fP6DIJ9gmMqMsCk6OjoEAslEckWYZg8GALMvo9XokSUIURQstknltvwX7uuGUTVWT5xOqKgBMXVdYa0Ccg4izljIvKKY5R4eHYYDGujFxmU1KeE+kNZUMPDXOAB6hFIKLGp1vB4SQGC+RIkKj8N5irEdgg8YvMwLHmRBFisnkFLtt8VKBcEgswkMSZ1irEFaFZ8rWaljyjYQHoahsjbA+TE7W4SjxQiKVQkYJ3hsqEzRMKSVaa6yBw8MDPv74Ez766COOjg7D5iHNO53+bMRP3n+P/mDA7Vu3eeutt7h96xZJkmKtoyoLpASlNFJG4C7SglJKvAPvHR5LWVUoHV9yyK6iytaRC32jwajmc15QDKxzFGXB2dkZo9GIPM9noF2W5az/tRbKxsYG/X6fNE1nWni7Ycqq+s6UM9/2j6YOvtGkRVAMEOdO1lU+Cu99uJYWeucmgJk4Wq07LP0Jd/jG8pw9HNmcs83vonPuevLagPu8KXhtDZXVgDsPyutqzO11y/K3vIz5ukpeVbndUsqy5OnTp+R5McuxIoS4AP4BZC1FUXJ6chKcikoSx3HghwcDelmPfr9H3OwnKoRACnk+QCBoQN7jrePs9BSPwzsXFuYUBcZaEMGsnZXRrXijtSHAmDqsYFWBKzcmx9RjrAWpNEIqlFLYeWftTFOU9PoD8qIiShTOGLxwswgbgLLIESJBiBitU8raEMcSIQVSK4TQaC2wzuG9wBrbVn+5eHj0+BFSx2xtbZJECqH0LL+MMwYnJdJo0iRFKMHh0RGffPIJH33wAScnJwwHA+7fv8c7d/d59Pgxh0fHVHWN9R6HYDRqVsd+8EFYHXvrNm+99Q77+3tkWdxQFTnvf/A+XsD+/v5MG3UmRGN4Z8mnU/LqlBs3boQ8OXMW2XX64zL+3DlHXVZUVcV0OuXk5ITj42PKMsT0IwNlsrm5yc2bN7l///4MvLvUYfcZy+jUZVRmy4F3Lc3ufq/zlgssc0B3rI/ZRLGwMWgnECHaq1ZNlnNWje/Wv33WVWUEeW3Avcu1zjtErpJux5unDV4WJNvnt1rVi0w68/VYRhO9aHmrxANSBO/9kydPmkEVeHBnL0+is0HU3OzxWGPJbU6RFxwfHs2ui+KYLE3p9/tkDaXTalFlMcXWZaA+PNRVSVHkRHGMrUqKoibp9dGRbupisc5fmESV1uRlTiQVWiuECSGBztV4W+MNWGPwKsE5cam/CCFRUYKKYoRKSNII6Wqc9LMFJc4JnDBUZYXUfZJMoeIdDo+e0u/vUlvD0dERg0FElAxQSmIij2cEqAvWx6Vv5OHBo895djAh6/Uoy5xf/qVfYXdrk1h7rKmQOGxd8pMPP+CnP/kph8eH9LKMt+/f4dd+9bvs7uwQRzHT6ZRvfO1dJtMJzw8O+fzRY54fHjLJC6x3OB/y2n/wwYd89NEnZFnK3t5Ok9jsLkkDki3VUZUlyjs8geOVKHCBv2+BblVk0yJrrf3ZTh5lWc4iSUajEVVVUVc1kdb0+30GgwFf+cpXZpahlPKC1rzOc+d/X/otxLkDc167b62llspdRJ1qrWf0TFOT2djx+MYaWVBnAd558CIkKQWQwWLCq0a3aAF9nnrxM19P8+fFgq+Q1wbc50O14JxzuyoqpTvbruPkfFHa5jJ4XJ1OuFtHuOzI/cLFQ1lVPH36hOl0wvHxCXEc8dlnnyGkZtAfzAaWECIsumi3q5MS3wB9l1sUogkly3OKPOf45Biab5AkCWmaIkUAbG8d0+kUgUcqxff/8A/5oz/6AVVtGWxusrExYGdnl5t7+9y4sYtSetbOSirSSOOtwZYVwlVIARKP8Mxi3qXoOoNDPaMopjcY0h/sYr3HOAcWLApUgnAOTw0i8NxRlFFZR1EGHl2IBOc1xnqOT04ZDO9RljWmtggc1lUkabTcKUkzXL2nqmqQBU+fHjCa1PQSyLYinClxtcHUNf1ewrtfeYtf3vh59m7skCYSIRSeYB4ooUBLNoYRG8NNvvSlt8mLnMOjMx4+eszTZweMxhOsceAs08mET6dnPHz4iL29ff7df/ffo9frhe/oG4O/+aSimcmFvOysbL93+/MC7dUAeVmW5Hketh08PQvx+iLEY2dZRr/fn1kEcRNp0p0ULvPLyxWe9vgyK7/7+6qggO6/1hc2v0F9t/wWh86fE9rMdwC5+9x2kgx7CZyXE/qEa3vHBX9Yl5cPZQkuN835Wo+r5LUB92Ua8Tpe/atA+0W47kX3dB21M0pDiIV1XOR0nC9rXXnZyaCuDePx2WwbuH/+L/4Z7/3sPQRBO+73B2xsDNna2mZza4fN4YBhluFljBBR0DxoN72QYZs52fleDY/qGiDPp1M2B32sMZR1TRRHfP7gc37v936Pp0+ehsGE4PjwKccHjzh43Cf+9rfZ30mwReAd4yRFiR6Hz56ytbEV2luKoKE3PJJoQgXBgXAIKVBxQraxzcbmLnGcAIJ8miOlJ5ZBm3a+h3MWU42RPmc8qajqCIREiArpBXs7+2hhSaKYL7/9LmnWZzwusT6Af5ELjo6fsb+3z7wpPS/GVigLWguUiEDGeOkQwgcqSSo2N7e4dWsf3UymxlmKoqasxmxubeKASe2J4z4CixI1/X5ELx1y79ZtaltzenbC08dPefzkOU5obt25y52797hxY59f+ZV/myRNg9MOCdZifE1Z5ownZ5yejEj722xtb1/yrXTB3BjDaDRiMplwdnY2i0AJzssBt2/dJUkSlL6sWXvf+ic68HQ+u1ySeeVp2dhcR4vvWuDtdd0yZ1q4P187053MhBCzdnHOYbEg5Gxi6G6n2F4zK7PRW6VvraG2FUTzewv2gX8/h4a2fy9+l6vktQH3boO2cpXGPi/LPv46VMq6DdYCPFy0MNap+7r0y4tSP4tECEFdV5yenhBFmn/8T/4xH3z4/mygGWuoyoKT4wNcXVJOT/l4PMU7UFFClg24ceMGG5sb9Hp90iQjAL1FCGaLLpRstAwhkEoxnUzRSlIWOf/kn/5jPvjgQ6q6ptVWpIBYK77y7pf5xV/4ebY2tqjrkqoswmRhoKoSnHXUtuLsbMqPfvzH7O/f4qvv3g+G8KyJAsDHScbm3k2itNecE8EcblaxAo1prJCqMb9t6xRzaB3Ri1N6mUIrQ10VYC2bwxgd1WRRRG0j8rImirfgtAgO4ys0qaC9V2itkFIhZsvTPUWR44Via2uLYloSRwmPnz7l6PiUu/duoVXIU+MReCkxVpFPDM5NSFLJVr+PqQriSLG3u8Pezg5f/8Y38TImSjOUilFSMx6P0UmEVoqzo2ccP3vItMxx3qG0wjnBNA8blrScdMs1B438FGMMQgiSJKHX6/HWW2+RZdnFJGi+dQi2/5ipo7PPdaFrC+Z7ehdw2z68THmaV8ra31eNtflVqMue2Tpv27UdXeu7PW7bhWgdyqe1bOE8pUGr4M9bCLCITrzYSKsUx1Xy2oA7rKY9urJqdl5ktq17/rrSDoB167nuBPIqfQZFkfP+h+/hvOcf/oPf4dNPPw3agD93AcVa8fPf/Ca/8PPfoqpKnPfUJjhVx5MJo9Exjx58QlXVCKVJkoStrSG3bt1mMBzw5NFjsjSm3+8TJSlJkpImKT/58Y/5V3/4Pc7GowCwzRZ0Wijeunub7/7iN7l96yZKBs03JNENDtDKWDbiiGjvBu+/9z4PP3/Cs2cnbG7tBIRuALtNEOZ92Ng6jvu0EQe+BXYkCINvgUeAd6JpA8kwjXAxpBEIYcBP8S5EyUjvcXXVTEwCKQWDRDDsaXa37yN82OVpoeY+w/BAzWgdB9oJEUDQS85Ox/Q3dsIydjul8JAXBWWVk6VxMOEbyksLUEKjhEFFMkyQcYypc4L2BwiJVgqdJlgJxlU4Z/nZ++8RfxpWsNbjI0wxob+1y3DzFjqOiWINUlBVFZ9//jmffPIJdV1z+/ZtbtwIawQuATnzY1RAs+ZA+kBJ+A6dF+TqZGeLxmgXzOcXKbaTzjxltEiWXdN1qM6Pufad58ucX4Q0H3p9oZxAygcqTIRoq3N65mpMexF5fcDd+4XmWfcjLuPpFtEy5yYgcAVPFw50HnqNtpzX4q8j57SG7x68fL4r16inaLRo7x3/4Hd+hyePHoeO2LjcPZ5eL+NX/61f5ue++m44ZgAJkbZkScL25jBojj4sjS6rmsl4HHKYHz3F5GccP3/Kp6Ow96jSmjTNGA6GmKri3u19jNnldDxiWkzppxnf/fnv8KW37xPFIZxRembOpqC0hI4vpAAhGU8KvBAM+hs081KjCYkL/KOY/fQ4D7auyCfHjCfTJqImhPqJxpfghcF6RxpJrHE4XwY+n7bvnIeizfqba6YgGxyzUzOZwdWiSf7G7i7jqeHZ8+eBrpB69p4eyPMCGZXBnpGSKE4Q7QVSIIVCqsZx20ZbCEmsI+RsJVZHayWY9hKLNWGRlZQCnKfMHc/LklRWDNKM/nCPs/GUo6PPePrsCZ8/fMDDhw958uQJRVHwjW98g7/yV/4Ku7u7szZYqMQs7JzBF9Isw1nVTdeW1rHZBdsrtfwLHPZlDb/7+yJ8WWSVz+eaaf1o3YiabnqDzhs0dWl/v+x/W6Slz9dt0XWL5PUBdwKfKxYsEOrOirAmvdFRJv51ybVm3daC5dVRMt77EP+rE9669xapjjg9PaXMC6qyZHOzx6/92V/l7p17WOuwzoOSHJ2NEd4S6QghRVj4kWiEh16U0stSbu5tU5YFaRZi4Z0x1HWNdY6yrBiPxlhTsr+3w1tvhfJNbRhuDEmzLIToGUcSNSvxfMsvBmQTBH5cEvKwjM7OODua8s5bew3uCVpfVTtYPB7clKooKcuCPJ9wenrEs6fP6Q2GYYFWkqBUSCmglUAKh3NlM+k1H0KA9yqAuFQIVIjOCYHh0PgbZOe7te3dFQF87ctf4qtf+SrjyYS8qBn0IiLpkS5MItZ6jk/OuH/vNhKBEgotFLjAjbetIfEoKXA+LNwSWiNVoxnPnsa5o9l5vK1w3uG8QgiJjHoIL/HOU5QFv//f/Tf8q+99n6PjY6wxCKVI0oTbd27z5S9/me/+4i+FSKeyCkvupcCLBfnTBeBbLrmhLkSzAgzTNk7bSqwzKBeNgXnN+Epf1gqI6HLp88daacG5S7m0TtJLWz3OUUHdsoIT2zdWZvtureOeS2Ws0xbrYOBrA+6tJrOMu15mpr2U+IszwBLj+mI9VzX+/M2i+4y5E0vOreX87T5qWYWbYqQQ/OJ3vsv9e3d58NmHGFPhjGM6mRJrwWDQD+yoFGgpqeuS2hhwhqquqU2NiiKGGxvkecHGoE+kBN4SImOyNIChrcPqykZj2d3eJssyNjYHFMUU8IH/jTSVsTjrSdIYIezMXL1suflZDL4QIYkXDXi1yk/75QL/bzg9ekLZhH3++Ec/46NPHlBVJft39vjOd75DomMePnzIzZs3uH/vNv0sCSDuZVOHiMpYTG2IIgVSUhtJmm7hrANp8eQ4Y85THa3gRL1zIAyDLGWQJShRIYXDmAAcWkc8PziZfbS2GYSQDWb6mVXbAr334BAzzfj8c59TIEVeg1Io1SzY8Q7R7ByFCn6jmztDfvk736AsCnQUk/X6bO3s0N/YAKUo8pz333+fKIoYDods72wz3NiYJSRr3zXUyZ1Hjszev8O5v+RwbTnwLv3SguxK31xbiwWXLLNE2nOtdMF8XttfZgkseAG6K3PbJrquL/CCc/qKe18bcAdmb7ws+uV6ESbtTTTa1sWyRHvuFXQ6Mftwc7K0up0Ht/V6QafJ4kpd/FMC0/Ep49Ex1gZQipRk0O+jAJWkSOdwdeCPnWUGKM6BagDF1AZTu8DPNhrZLKpACHq9TZy3IaGWV2SJbnZGEjO6oKgqhIxIkhgvaNyr553+4icRs8/nHKhIMxj0wlHhLlztvceZmg8+ep8/+sFPePDkCVpp3r5/h3ff/QpJEuERfPzhR0ynU27t30BrhRceT4wTnrr2lEWFF5ZIaXSwo6hMhXMDytIjZIyKPEIURFoTRwpbTFj+scWMRwrzgMFg8I3fwCMYT6YhhpqwqEg1i8ZCHhs3syqEVCih0doxH/58rr+H73E6rqlcyc7ONlmaUORn2DpHKx16hHDs722xv7cNQFmUVLXn8Cj0kcHGJjIKJFVVVRwcHHBweEicxHzlK19he3v7/PneI5sVXZfXcYZY7vPuvZ7W3mrpy1JxLwL2RZbT0mdw0bm5jiyigLrU6vnjfWsAXn6qmKmxzPeZVY7j7nPnjy+T1wLcgxa22hxZNYPNX3tBGq3hgtZ/4fzCRy+uZ2fSuWRFtBrmekVduu6i8+V6wH5Z3+3W6zwmXHgQToB3eGEbflnMlE/RDFDXaHpSiMbF6QGHd+3y93ORQqKkwlmDlBGHxyNirdjb3Ua4Eu8tAolWMdZZnANjqrCgRjQaCBcHTbcNBBBFknv3bvPOW++wt5uBr5g5EDtinac/2ODb3/4GX/val9ne3qasSn723gdMJjnf/aVf4s/92p8hinTgLwi8flEYKlujZVgMlfVCCCXCNRYDwSmApKocwlk8JYKK/Zt72GK8kJu9+DVmOlvni4WVvd638czBR5KmKVkVcqp4Z3FphvMeax2IkJpWConWwUF83lJNPUXYyk97gxcCqaLm+zviSDUUysVa+cbX8Oknn/Dhp5/RH27wpS+/y9tffrezwEhRliXGmAvrSpxzs+8lFyhgvrXMOn9fpah1AX4V/TH/e/eaZSAbmlrMAL69p/tzFe2xDGtm+uRKpW6xdLn1lVTTNWSdPVTvA/85cIswon7be/9/FELsAH8XeAf4BPiL3vvj5p6/DvxlwkT+17z3y/NScv7KnpC2VaygYK5Nz8hznfoifXKO6utSMedaOlzoMtdo+0v1fXmL9QLvPPv/OX4EXxyhXWcxtU3nF7igHePRKpqVE75DmAjwwUHp8XjhmoDdoCU6B9aeHxfCYZzFCYEiaOtCuFmHF0jqqgomvJcNLdNpn9az6j3SC4Qw/NJ3vtbkQvFIDK6BouBdbe8MZfV7Gf1eBt4ynkzYGPb5pX/rF9jc2mTQ38JbhzM17fITiUAnMRgJPkbKGiGbyU42hTdhnr4x90I+mnbvgIuffz5qA0BJjXNtJpOm3ZsskEIyY9Y9AudqJvmEPJ8G8BES2fQRJaOQVdKVBDrYk+nud2/7c8gH7pvvgw/bBma9Pvl0QpbGdDte60L2wvHVr32F2/fu8Mc//DEPPv+cXn/IjZs3SbO06QqeZ8+esbW1Rb/fv/S+3f0V4BzI5zXsRXTKbHwtnCQvtu+V4Ot906ahnbtfqVVmzlvufFw637qArzsqz/vt+STj5vS0i9p9t12u41NcZ3KE9TR3A/zvvfffE0IMgT8QQvx3wP8G+B3v/d8UQvwW8FvAbwohvgn8JeBbwB3gHwghvuZXbJTtnKWsSqSKQzIl2na6PEu/Eq79AvpdceklYL8E7WuXcdWx68gMVDrz1TIezwkXsioG8rXzHp7zCAyPIIBVmvaojES4Gq9MY0R2dfZzCsULTZINqesILzR5XqNjhSOE6knvwAeuV4hgK9g2ttxDN7GN7HINMzXIEEmL9+Z8+TYt0LYOqW5jtEvCJYPBACFC/hmlelga8FYWicQ1m2wIqZDS45wMESrS4b087yYy+BiC8uHQQqCiAJ7I1RYmCKI4xUtFVRSBg2+nFuHZ3hnyVnUr0Cwu5OcZbgzJ0ozN4SCs9HUW7yyxdnhRI4cgVITzbvZVZtZjW2mpsHWzSKpJjyyVatqqO5s2YAFEccRwY8D+rdtoHfHxZ5/zox99n+n3anZ297h16xb7+/uzBTuLNOv5HZ9aEJ8HpJYS6Spr84ETi6JE5u9v71sF9N5fBPf2s85/J+/Pp4NFk8w6O1hdtB6Wa+DzlsKiFOTn5Vw/PQmst0H2Y+Bx8/tICPET4C7wF4A/31z2t4F/BPxmc/zveO9L4GMhxAfArwD/ctkzptMpH3/8Mds7e6RZjziKieOIuN1JZvY/ln7sL0K6WoKUnef6czfWRRw9N7lnZmFXYxBdk/xlKnaxfudlX5wAhRBYa3j+/FmzRdflclrAdNbjpcc7jzUCW4PWHh3pJlVq+B5KqxnY9wbbDDbblAEB+L7y7oCyqojiBOnA2xgnKjzVTD8NWouY7SXZgvLFql0GzNkR34K57x7t9Ak/O+ebKA6lezx+fMDdW9soKXFehMEnQMzyjTROWyEuLDQKc2JwFjvnUJEm3RhSFdWFerZ16Fp4zmvK2gMmgHGDraKZkIbDPhvDTbwpiJWkl8b00p1wka/POWFvseUZXoxRQoBT6GbT7IuTeWhLKQW1sVR1hRBp8x7tlN5pt9mkGKw1qRRRknDz9j7bN28wmUz5R//knzOZTBiNRrOEXq3WfqlLzYFSdwxdah9xMfJm0SQ5X0573dqOTC5ateH6lZeHPtBq4p1+GRbqhVS97fqDeWmvDt9ZND8vg/yyd23f77zul/vXouOL5FqcuxDiHeC7wO8B+w3w471/LIS42Vx2F/jdzm0PmmPzZf068OsA/X6P4+MjrJMMhyZk+LOWSGu2NjfpD3oodf2EYOvwZyveda4saLc6w7vZ7H4RrOedP775uMvzyXTLWKuGrYLWAfDZ0+c0Ke89k/GE3//930eYnK1hr1Oz839BM3RIfQ5VxtZU1Ri8DQt4pKYsK4yxREohpGS4sUMcBz7Y4RFekvYUSRZWseJjnPUwHoMJu9hIFXjiujJoHXFB+770rjMuZIFc7ORaa4w93ybN+wBwreNSCYmpSoTwWBsW9jCjjkTY/csKQqIm15lAz62k2dQhBZVwOLFwfHc+lWQyqamdxuJwTNjZ2UIJhzAW32zVhg/5aoR0eFc22mOTVli0oNzy1k38p7U4W80/EBotvNdXxL0t8BLv2kiWVU3pMDZkvJwUBSdnp3gvuXnzFn/1r/41bt2+w2A4nG0qvYzrvlCs7yTXWuCTCN9o8Z4Lq8C7q/EvosG8byevVSNq0boU33z/Fcv+/Zzl00pHF5gdmusbqzFgfZxa1+pfG9yFEAPg/wX877z3ZysesOjEpRp7738b+G2Amzf3/Pb2Fl4YxpPnJPEQpWKqquLx4yf0eimbWxv0er0L+0Ium9UW/S4aSuJiNS8C8bmp347o+XAjefENhbjwc6mpd+lEOyguXrvgkoWnV2lM8+cm0ykHB0fc2dvoHPXzUxKRDjQFeG7s9HAuoq4z6qrCurCTUinlLGlX0GzEHGh0ILC5Rsx48dDetjYU0ykbgz5adgfgubYU2r8pRoBQcbNZR1t3f2HoCkGIAmrAbVaelyEaRkiQkiQNYZtShOgevMETnzfu7PO2S83DwiKtFQ6FUjV1bTkdnVKMJvSzjJ3BYFbv9tkt8DjvCAuyBEVZUpsRg36Ct46Nfo/atlq3DNu2pQlOOCSmqZMGPFLJQCUp1VhINuSxnzV384uAluZyZoIQDkFMWdR43yS/WgR4AtroIyEjhIipastkMuXtdzb4ylfeJU7TS/1rvq+1OZfOrd3F+x8AF/YsnT+/iI5tj59P3PISqF96p1lXFJfH06JbZkrGvFbtZ7N86zu5jrR9yV+yNi/ikW/Aoq3vIq3/OszFWuAuhIgIwP5/997/v5vDT4UQtxut/TbwrDn+ALjfuf0e8GhV+c45njx5Ql7m6BgindDvD+ln2wgvOTk54tHjh2xsbLCxsTFz5rRaxCJpG3LWCN42MdiGqqoxxs5M2hAJED5er5cRx0nIkOcuduRZZ1uyvZb38xDdTkDNwGkGcqMvt3fNrl0kovvbmppAW09jDAcHB+zv7yN9wXwPb6ckJR3eOkydY10NVAjvSbQkjdJwl/T0bm4jhAwLXgRYU1BWE4o8xyHQUYRSEUpqlIrC/qIiOG3bR6dpzK39XYaDLGiozs1A6dwXIDrd30MdLJ9JXhEnGinOh2AbiSGkANNpE6mQKqY36COEBqXJBkM8GoiQOJxQDSyKJqpEorTE+wCCEtFQMQIvLDoB6Q1VNUVpSVEVVM6GjTea9+gCkvMO33iQlIpAxEghMc4hhUZKOBuN8F6xubXJtDJkWUplPZNpETR259jc3MR6SW4kSvWCY9lPiCKFxGKqkg6Sce5REUA9GwvdDIUX+kH4wHgcW1s3iNINrBSMzs6CL2CBInWpjDlacJ5L7/bNRVTLVWV2uWnfRuf4DvjOUXTNjefvN2dRB2xg7p4WgNtnt9c1ZQvJpRDcy09d9jaddmnb4eLz/YV2IjAE7txVuSrVySJZJ1pGAP9X4Cfe+7/VOfVfAf8J8Debn/9l5/h/IYT4WwSH6leB31/1DKU0Ozv71MY3OcEjjC2pioI8L7DWkqYZZVny4MEDnj59ymCQMRxusLOzTb83QClN+3G899iqpqxLimLK6ekJ5XRCnucYUxNFEVVZUZRlsxtR2BSisjVJknDz5k32b+5zY3ePOEmgAYF1zaFO47W/AC21A7NOc37h4tsv/BK60TITt5Wu1qOUIs9zfvLjn/AL3/xyp8SL95k6p/HvderVtiUIIdEixCtbZxDeIhFMR4ccHp7w4YcfgZD0+wN6/R5pmpFkKUmakMQRztQzAE9TRSYUYJrJ8HJY3sWfkrOi4sP3P+f09Jhf/dVfQOr5VczNtVKgopQ4zegNNkjSPipJyPMCbwU7Owk4i1MaXA9ZTxGY2UC21jR5+13jYzkf3M5bZCKROLAepYJzOHcVSaLxRX1hKcU5kIZwUykkXqjZgBZCILzk6PAMEfXYuZFSF2dMneO0zDk6POCtO7dnNFariToL07FBRwXGwUavR+3PQw3PwaaxPLu9bEXu+XBtSJWcZT3S/n32bwZH7kX6cbEmOZ8md34T9auiP+b/7tItrYY/mxg8Lfo101jr6JTtm18sb/FTu28xO3JhIqK14EWzBd85vVVUOR5J2u5cdV11foEEw6F5b0Fw7S2hvuZ5+0Wyjub+a8D/GvhjIcT3m2P/BwKo/z0hxF8GPgP+o+ahPxJC/D3gx4RIm7+6KlIGwLkaa8/wPqKuoa4mICBLewyHm4xHE05OTplMpkynE5QSHB6GjWqjKGJjsMnN3ZtIKTk+PWU8GjEen4YBKkUAdCVnKTuLImc6GTPNp0wmAfTDruUJppfxs6MD3v/JT7h96xabWzsMNje5d+8+cdqbfYBFXN/yKbw1Cy9HnMxftvj38Ed3hd68LHOs/tu//Mv84s//Ag8/fY+Hn30897zW4SPwQuEwSK2xLsH6GkFY8n7v/tts7+xhraUsC4p8ymQy4uGjh0RpzP0vvUVdht11Dg+fUxtDXQdguHNrj7t3bod6NxqLvEBnnWucvgWx5n18E8FjHJyMzhBehqyKC/lQ0HHCzZ1b6DQmxOxLHB7nG3pAhNlVEhYHYatg0c0iOhpAFAEIhQ+gHGmFA0zD4RtnsLVFSCjqnFhEs3qffx/RWCPtcYmzYI0L8eoEesgYT1mHfPeIsFnF4MYWMtEgdcO1eyRNaKgXCO/RSqNEh7NuG6GZOHr9Ifk0D7y8bDb2dotpGd/5pcVN0fiP2rBTwUUKrusU7QJ8ty/C5QiTLrXS3QGpq9F3MzC27dmGWEopZztZBVBvLL2ZAnTeH9q/531j59p5l8W7nAog/BLCSoX3szBJ52p++NM/wscJP/+Vr5NGgdoTl3j8Tv+eb/OFY/jiO4RqX2zbbltdJetEy/wzlsPWv7/knr8B/I0rn96IsVOOjn+IVBlZ0iNOeoBkNNYoHaNEytZmgtYxbmuT6TRkKjw7OcGYmtPeMc8fP6Eocqyp0ZGmqkuss/SHG8FBWzu8t9gm57izBiUlSRSRxTGnZ2dMx1POTk/wzeDJp1Ok+oRev8+TRw95+0tfYWd3l6zXR0kNsnVWLdI+ml8Es0HHXAc8v+CccpltmuFCfHmribVpReF6s7iUkiRNmpWJCzqEB2MVp2cjoijGS83TZwUff/YeUtbcvLHH4cmUw6N/ibE1Wifc2N3l5q1b3Lr/NZJIMRmPqYsJxlSYugq0DR6tBINBSlXWOGsRPoBjVTvOTkdY59jf356DjXkRKNWEvV0IF7usn0kpiePeQvKpfdcLdzQ0kGjIE6UVUSKIk3C+riq8NcQKpBbkwoOwYds3rbHOcnY6ItvaItU68OBhQcBMe56FKgJVVVNWIf3DTCOTkmpah0y5AipnGU8rdKQJs2HQHoWUs3j7AGYSKdt+1YJIq4F68rwMNtHMQQh1WV7YlXO+nfEwHY9xPvD/SkXQpC24cOUS3r3bH+f/da+ZB/ZuWd1Mq90yL0wkHforzESXe885fz578tzvHeKvU1T7TNFBfucsRVWSphl4z5PDZzw/O8BHikcHm3zp9ttoGcJMxUxhai2OTr6ZpmNYa5kWU3ppr0NANvVqJ572PTpj/jqRMvCarFD1TpHnIc9IXRWoUYhq8ILAT/oEZxRVaXFWhBzJws2oh2I6ZlQdU1UVGxsDpFSkaUySJIxGYzyQxjGls0yn07AJsIe6KBHeU5uaQa/PGdNZzu3pdMro4ACAmzdv8uzJY8ajMzY2NxlubLK1tcPO7h79/iAslLrU4F2NvjODz8B+dpIZFHQ6cJtTW3SOt9fARdN11Ye31nJ09JzxZDJXv3PzLy8NRTklTlIcFqVipISimDKZjHn8+BEgGI3HaK05fPaQTz76aZOf5T55UWCtZ3Nrk6Q/QOsIhSOJm9SmZ6dURQHOU1WGk/GE0dkpw8EA4RUIO58dotMugbvO8xxXVzgLSs5WQpxf2rS5lzRZJoPGaW3QlNtcjAIxC8DxIiyMSpULGqorQVuwrtGELd4ZnHHY6gSXKpw0RNH5RtKmrqnxRJFElq0LNICwFJ6NfooloTQG57PgbbH2HDCF4Gw0xeKxAp6PTjkrp2RJyp2dG03AkAjg32xq7aQALfEqUFHtkptGxwMhOB1ZkJbhRobCYasSiUNHCilCSrLZlNpYSHjPT374A2pn2d27webWBhsbm8C7+A7Az4P4/PH5a7q5WVoAX6Z5zjtPu7Hyl31c4gJ+L3PeXrzHd3622vAKSxjL4fEzPvzwQ7757Z9HCsEnTz7BaEdVlTw4fMjtvdtsJjEW21mr4S9QVSF0MpT75OARP/noPX7527/CMDsPKQ1f5GKww/xEuiqKaF5eC3AHibMRSqVIkVLmJWVRIqUMS8WlpjYFtnGY9foZVRV2gBmNxiRxzObGJsPhgOFwQG0Mpa2ZTCZUzhJrjRCQ5zlFUVCWJZHSZFlKWZZ4PHGSkjmoTUWSJHjv2d3e5s6du5RlwenpGc+fP2MyGWPKnIMnjyhrw7233uLLX/kqvd4AoZrQPvz5bkFz9I1vtHFrDUo1DmF/DlTBqRhESTnjAedBvf29lWWajjE1ZVFySV8T5wuyIh2RphltFKCUkv5giPNT4iRiMOgRRTHDjT43dm9wa/8muBI89JIMiaQ2U7x3/PTHP8Y5x7tf/Sqnp9VMc03728RJDx1JbHQKUpE0C4HEzBnhOR9o5/Wt65raVJiyoqoDuJ5LiM1vuVjhHd7U1LaiLHJMPWUyzWfwF8VRs3NOjHQG4Tymys/T9jb/P4+ckhAriGLOJiOOj4+Y5iVCnFtZxm2S7d6A2oTFWs1isICXBVCRaMXtG4MAqkJhjA2rgAXUdVix6wWgBL1+bwYSYU/RUBcpBbaJnT6f+9q+4S9glHMeh6WqKob9HrYqUEqjZJPrpZ0c55iDNE04ePSIg8Pn3L59h5/7uX4TD7DYuloFNrMInU7/X9eJ2t5/qXx/fp2UkpAv8zLru8wftaieLWW4iN8uq5L333+f45MjPv38U/b3b2KcwVnP1mCbftonn07py5iffPw+7779JbIsm42tC74D7xlPx/z0vffI65yj40MG2fm39r7b69ev/zJ5LcDdO7CFwFISWUkmJcNhHyEUeVHgrGHQG8y0sDRNqLTEGENv7wb9QR/rHNZbwDGyOaOioJhOiKII6zwPHj7EGkO/36fX62GsZeRChr7dvV2EkFSm4ujoFGtrer0e4/GITz75iHfeeYdbt25ydHhImReMz864dfs21jqePnzAowcP+dKX3+X+218m6/URKuQpX6RJOOc4G4344Y//mG9+45tsb26DV8yiRcLIC4NJQFGFySb0vYavvmKgXNLi3Tl0XXDENgAvpW54TsJ4F4JeOqCuT1BKECchLYHCh6iioGeCACcqnA8bPeM8J0fH5PmEb3796wzSIZaK0XhCOR1xcnwQrAKl6fc3GA76xFFEJB3GFsgqTLT4znoBD1Vl6Pcz+jt7jeP8ouXTXuusYXT0OaaqqcuSo+MTHjx+xPPnh2xtbXPnzh16vYyqrkiSmOEgI0ticM0mISKwpueh9QJQeOcxtQ1x9F7Qy/oYa1AqXFgXJbY2IbSzycIYrC7R8NUeXFhhG1hbiWu2LpQS+v0UKTVnpmRS5iAFg/4QpTSSZuWu9yEVQRSRJAKlapQmxOd7OtaIQ3qF1FGwbkXLLXviNGMyGZFmCeoyIQCE/PP33nqb2hRIHNLbzoKu8/61yFLsrkadv25RP10GwO2/+XzoVVUxHo/Z2dlp6Io21Hj9CJJZfdoh0G6ZOVcn16wkfvLkCScnJyE18+gUH3uePntKbiqstQx6fbywfPLkUz5+8BFWOr757jeIRFgLQkcZc7XjybOnjKaBkvzk88/Y2tphe9CGKc9vzH2u7Kzbpl15LcBdSoiUwxqJEwKnNRWglaS3tYFqNtSt66DtOmdRUmGFZTqdcHBwgNQhF/WknDJxNUnWI0lTbBmyF25sbkCzJVwURWR7m4xFxaZKkZXHScHG5pAsjnly8JzxeMzmcIMkSfjss89wzpHGCYN+n7KsmE6mRFFEWZYcn55xenLKxx9/xDtf+QrvvvvVoF021N65AydQDO+9/x6Pnj5Gacl3v/Nd0rjHbKZueT8pODg65Cc//Qnf/ta36ff7PH36hHt37qKaUMxFqpQxhucHz7i1fxsInfT4+Cnl5PQisMNMVZBSoubCSpO0R1Ilge/2KiQH8zaAa6v1+2Zla+NT8ECchM0mRMtBAkqAMyX9SJD2M84mBWdnR0wmJ2glG99HjyzbZncvwZiSosxDGc6xu7PDn/0zv0oWx0hhmvCwRnX15+1rreXJw894//0P+fyzhxR1zebOFvfv3ePWjRtsDPs8eviQzz79jC+/+w4b/bcAFygcL0JKAjTGtQ48EFJTO4sTCqU0aZZxfHzKndu3kFqGmHlnoKjC/qyCxm/bCQeckbnMnMqhPzSUeCQovCE3JUIKdBQ3W7gZaudJdITHY43B+YqqNKjIhTm7NXoaB24rbV+zDb0fqPrF0SSzb57ERHFMfzDgbGzB1J1w4C5H7S/x5aE7XQaerqO0e7yt4yIt+vy6wIfnxRTnHA8fPuTBg4f84ne+w97eXrNQvFktukTm6+MaH5mQ7YK1i7RHt05VVZFlGdvb2xwcH1HbmseHz4gHGdSasir57PFnFFVOOc3Jfc7Hjz+hN+jzpTtvEws1K9/7sKXl2dkp48kYpTRRHDGZTuknaVi/M2dFhWpctobao1dNZ68FuINC6AGVLXCmwsaC2AukiMknBUJ46rrEVM3CDXzQNrUgjiOm0wnGwTBNIdHYvG60LDCNSdnLUqhqrLXoQYLra3xRMdEW5R21AlsVFNMzrA8f1tnTZmehlDwvODw85s6d2wyHGxwcH6OkZDwZM81zIq0p8ynv/eiHHD9/xre//W02NrcDX4oIQCUEZV1wUp4iUnh2+pSHzx7ypbtfQkmNb7xqHqjqmvc/+RnPz57y/Pgmo2nCH//kj9Facmf/NmHpkQzw2S5c8Z7PH33OJw8/ZnNzg2FvA1tV/PTHP6QfCTY2hheAJfziESpEmgg8qrYkWYKIJNbvoZUhaTq89R4dxyAUECG0Jk6HZP3tkANeCu699RbT6RQdhS3iJGrmUvRNAnQtI7LUkyUx/SxGeIMxjrPTAyaTAhUrknRAlikGwyFJklNXoW/4ZqsoT0gXEEqWSCeprSWONF965y3eun+XXn+AUIKz0zOKYsL2ziZf/7mf41vf+FoIWRaB5bQ+obYhuVlVVnhfESlBmmqc8JQKiMEUhuOTo2AZFAW1DZt/60jjN3rYwqKcm4HNDKgW9PgA7IKbt3aIb2yRVyEdbzxIgrbjHLk39JMIL4JFmkQyZNWUFuHAG4+PHJ6OleA9TgYtv6yqoD2Khu8VsrnKddwUjQaCD+PPCczZmIPDI6QXfOmdmyA088izTHu8Did8VdRHu5r4Jx//iJPJGDMpqUrDw4cPuHFjBzGPhgvqsfAZXgZrGYkQIQncZaevIc/HfPDxBxycPEdEimyQ8v6nHzAtJ/QHAwaNf29a5aBBZxHjYszD40fc3tsninozS9h7xyQfcTo5xXiL81DUBXk9pbQDpBYozieDWV064Z3BSmmjozysmNTgtQF3j5Ce7WEP35eMXUUqE8pJzjQvGorDz9b9VFVw4KVZTNbvc+fOXVQSI7TicHJKVVZooWeacxjEECcx2jvKuqCcVFTOIKUgSWJGkxFxGrF1a4+TR8cA1FXNNM8Zj8fcunULJRUnJ6copdne2gJCeFRZG6wEH0m8dTw7es4//91/wb17b3Pnzh10FKGjBBkpnpw8ZWpzdD/G1Y7xZExdV8hYXwh6OD075fD4CCHB+ppHTw4YT8c8ffqUW3v7QYOeOZSC8T/Ox7z3+YcgHeNqSr83ZDjc5D/8D//nPH/4KZ99/smldocQiaGkDAt2vKMoSpwTKBnWGwwGg5CoCU+SBn+EkrC9e4ukN8Q5T9yMjXe/tn3uCG4ckkxzwM40fK01yoel3g1tTBQpsiwOjnTvMOWY4yN49hSklmxubjLobZKkYRLNc4N1dTMVnmvxbR4cCFy9VIq9vT36g35wGBuHt+V5Q6OojcB4h2zWM8RxjBIzQiVYjcYwGo/Ah1QLzw+ezx473ByyMRxAFoGxiNLgO5E9WRpWsRbFpFXrmwyLjixJUJli6h3TyYTSmBn9Nuz3Ge7fDLsqOQt+giRncxjhhUKokGr5ogSyP8siCqfoqu4tfHVpiC7sOWdxTiJk4Jqno4Kv/Vx64ULnF2+QsQ6PfkFTnzksO8XPivCzQILDkyOOpqdYb1E9zUbaY1rkGGOIdcQV88eF57bPts5y+PyAvZvt/r2hPr6tlw/vkxdTjk+OcN7R7w0YT8cMNoZsp1sYYzk9PaUoCgaDQchv35httamoTIlTaZPSOViCVVXx9jtvcTw+RSpNaSs++PRD8J77t+6idCDLutp5O6m2rzkLHW0Hzgp5LcBdADhLZWqMURS+prI1wzhBOIWKY0bPj5mcjCnKgjiO2L91i53tLUbjMWdnp/SHG9hYYZ1FK9ksz5ZNJJjHS0E27FHjcAoENTESGSkq4YniKIQb+ojtvRuUkwnlNA+70/swoThnqI3h5PQYrUPWwTiOGQ6HiDSmtIajgwOyKGbQ6/PTH/+Qjz98H6UFXnqifoZNJaUtcbUhVTHWGMq6RimLbOKirXUcHT9vHImGw6MDjo6Pcc5QFFPKvKA/iAJINCBV25qHTz5nND1BJZJJeYZxN3DG8LOPPiCVy/PbIJo9Q2QA+lhLjAErY4rK4pzH2KqxZixCKTzNTx/AOOB2QDspZTPxhIU7iM6CFnGeKkwSUglL/AWNJFBniq3trRBBVeeU+QEPDkuM9SGx1a19hkiKYoqwNecbRDfPAXQcHNwIidcJVkRIZRG+DvDR5LTXWiO8C8CGRChP612WApSS5HWNkGF7u6iZEHUDDMZYXF4iZNRYK+e0CFJQmhKBnOWikT4oM14IXENLnR0eYV3IkaObBUDOWlxt8CpC+JAEDu+xVOG96vNPOFt6JtsdnMbs9MN+sXVZhG/URgEtG4deoaMEHSdkWYYUUcgb5D2tt2A+Nh3Oufb5UN2F/a2zKXTnc3XopeZPD9Z7Hh0/4/npEVopBmmfuqzY2Nxf8RYr3q9R8iaTM9774EeUJufenfsIAZPpiDTNUD60fVlUHB4egLNEcYTBEcUJ9bji4OAIawyD4QY3tneIoghThVTNQkqKsmRaFmzHG+ffRQiSJOHg4UGzh0CBKEviJObw6IAb2zvEUXKB0gvN0k7EokmlEc6rpeTaubwW4I4AFWucdFgcxobNAKrEUwuPUh56CYlx7O3f5M6d26RpghCCaZ5TWUOhoa5LEII0SUIUhAnAJ5XE4ynrCis8VqoAaEqitWY6nTJbCiJolo3DYBC2iLPWsrGxR51UnJ6c4IyhKAuqqgqx1WmGkwLnodfvQ2UoyxKtNUUR9iY1zjIQHhVH1HWBrSxpL+bJk8fko4JYp0RRjDUhcuLw5Bi8RyvF4dEhk8kErTSHxwd8/0ffI44Skjhh0B+SZX2iNObp0TNEJPDC8fmjz9jp7/P08SOeHh+wkerLE31zQEpBFCWYOqewhs1+s18kPSpzu1kAA4OkR5aGXZsEAuVBhITuhAx4ouHjaeio0Dl9SxA2nGakFUqJBsTaSachz72bdeA2/jzSkqjfY9hPGw625vMPPqQ0hq3dG+zt7pJlKWI6IbdHOGvxNjxUNKghZMbB4TH72wOEjjFeINy0yV0jOh42gZSttyBo186GxUxZmob8Lg6iSBOpsFFJbWtIIpzSqMrgbdjoRDg4OhpjnCLNeiE8N43BtrlhHKL21KKivzEksZ68nBIrHTJPRhGldwzOsZU5XXeOlWhik1xzg2vSLc/UW08beXOhD/j2bsH2zg3iJFh8VVWh42h29bIQxoPDQ6IoYnNzcyXkhGiXi2zxbLK/cKMA4cjzKYfHh8RRWNRk6gpT1hwcPufo+Iibu3uXJpL5qJzL4YOOx08fczg+Jn72mJt7+1RVwR/88F/ytXd/njvbt6jLmoPnz3n69AlCCKzzKCECTijBjd1dnHPESUochxxYznuSNKWuKqbTCWfjMbf6N2YhvtPplA8//JAHDx9iRAjhjmPdWIeOw6NDhBNs9DcuTYrhs5+bXt77Zp+KfwM0d+c8JTVqmCC9IdVpAFgd2MG6LOlt9xiNznj24Qd8+umnzfLwoB1t7+0SaUj6PVwFxtRoGVYYejzOghGe2nhUFuF9yIIopMI5SxJHFGVJldfoQUJ/u08kJadHp6T9AZGOGE9zNocbpL0ejx89YjQac3P3BkmaIOKICSbsoiMzqrMxkZJUZYFUiqKu0b0EHylqE3bR8coHq6G2HI2eIa1By5iyLBFaYVEY7xCq2QlIhO3evPe4CRSmojoaY/KaLM24eW+fkS3DIhkMZuj57MnHfPbhR0HbSjIuOI84d4qKehoAwBpwNQ7TRHOAEg6cRHpJFklEUwchoSpOcLnn5PgECBNl2EhYoaIo5GpRwSka4qgdUZOjJYB9k2u81U1EC6ptBEgAZyVSrDdNrR1prNnZHFLZmqoY8fknpzgvSdI+e7f2mwVoZxTjs6ApNouBhA9cPU4iRIVrVmEKIVEWEIGSEMKBkDjpMalGCIk0NbGP8VJSTAviKJ5Fyyg01nucFggU3oAygZo6Pp1wcmpR0ZjhpuLWrX2wno1Bn8qO8ErhhaIcnTHNc6yzTI2lLEtu376FHAyYKIFMEqRRSJniMTibB6B0Fu/tOWh2qDog7MHaWAxhobiHOVAQtOGVYWcqqSN6UUzWF1hT4USIv24hcha/LQSPnjziez/4Pns3bvDdX/glorkN7oOEyUYIT2UsKtKodgPt2RzlL8xXlbF8/OhTDk4OKEyOlIqt/iZJElNUBY+fPw7rKnQ8ozKEaFe2nvs9Zu/YgLzzlpPJCR6YllNGk1MOT59zNDnl+z/9Q3rf/bPELgQPWG/xEpJego9EoHprOBuNQQh6UmKb1L860mghGPZ7SKkpyyl5OaWX9fDOcXhyyPHZMV4GP6H3wTqI45iyLrF4pI4ag7edcD11XXBwfMT+3s0wNpSf4ZoxNavk9QB379CxonI1Z+MRURwTpwkIT5JG2EYjHG5vMiRhOBgQxTFlVVHXFWiJSBOmRY6tDINeH6zFOo/1DilViBF2DiUEKonRLnwQ5y2T6YS44cStNaBj0iwh3b/Fk6ePMd41GftAScWd27c5PT6ZxfCKZuVgbcOk4pwlL0v6vT6mNgx6KenOBmQx4/yMJE0oRYWlBW+JJuH46AwvHP1hH2TYsxQPQoVO5L1HRAqLI0kjfKKRBrJ+ihUOY2v6/T4oz9lkhKkshSnJogSl1YJoYACB9xY50wpDjLabmYVt3H6TK2WmPQgOD55R1zVPHj8Ji2tcm8RLgoQojtkYbrKxsdmsjBQhOYr00IB9oGHmAMeHDt5yoIWtOTo5Y3x2xt27+8QqaHZSQBJHbG/20EozmuR88LOfoJRie/cGe7fuga0ZT/JgYSUpOzf2UVHE2dkRxdkBbYZJOm9LqB3K+hCDr0JfMU6jbEhcFbaa00RRRJIkRFqBcRjpicS5FVPXhryoiYVARTFehBw2XnjQirNpQVEZ4ixEqozGI0ajEUpIJmdjBjpBxRnCWwoTEpHVlcK7Kmxw3k8p8+lsYmy8KCgdY+qSEGIbLDFr65n1dAHfO7+f5ypXeHu+0t8TVkm3OWOccxyfHPKDH3+fcXGCeZZz+/lt7u3fubjpSvsIETIqfvTZJ+zu73Nzc/vCs4PD0Df9EWprGedj+sMesYtIkwzlJK4yCODZ0TNuj25zY3sP2WysMmMGV8hoMubo+DBw9mmCl57xdITFM8rHPHj+iK/e+zJxGmO9R2iJ1AorYZpPOTw8oshzpJYYZ0O7GoPSulkfY8myjK/cfocsywCY5GOcN9zYv8Ho4YTJdEpRFEgk1tiQn0aGNRjPnz/nxo1dvA/W4cnpCT/74H3iKCJNEiKtOTk5prSBPVglrwW4B8dkMO2TJKauDdOJJetlxFrhlcRYS9JLSIjBeMbjMZHWpEmKSmMq4ehFCSLrIzzUxuJhlgDKW4+VAuMsZVWHQeZN4xzUWGOxxiIdlLbATEoymbGxsUFl6qABAXlRIKQg3t7g6OiEm1mKkB6JJFLBhI2HPXTt6CVpcBoqiYsUuavDFnRNMjRrLL2sh1MK5yWb21sUZcmoKKmdZTgc4AgmndY6ZLV0Fu8UpvZEaYISMTJLmVQFpanQKlA5TjhqX7G1s8Gd7bdw5YSTUX6p93vAC0mFCEnBdExVxozyI7QWJHGEkrKZtNqojAAAUig2hhlJlM5CzMDjbJMbHNjcDDG8zpjmrGi2ivMY54m1Xzggw6Kk4Lwcjaf84fd+gBSee/duX+Axwy8gFfR6EXt+gHOO6eiIHz15RJz2uHv/HYabmyAVOk2JkgQxKUIaXwwtYz1jLKTACxn8AcbjK8vB0SFJGlOXFf1+nziOmU7HHB4ekiRB4RjEcYjysg5pAz/qXHC6h/bSszYXAkTtkN6TDXsYZyiKAgfs7OygG0dcaQ27BMUEocE7rLFoLWjXRnSnRuGD9h0nIZe+cyHdr28cfcKr7hx9qS+cnTwnzyfEUYqSIRRW+E75M5rDM5qMGU3PMLYmLy1nk2Nqs0eikznjQCC95ODoGT/86R9zPz9j9zu/hGxSHQflq0nh2/aLOOL2/j7PTp9jK8Px+JBEpWRpxiTPcQmcjEdsDraJdUjihg/0ydJIHut4/OgReZGjteLg8HlYQCYccZJRm5pHzx5hS0OmwzfNy2kIYZxMOD47QSDZ3NwiSVOSNCKJY3BQ1RWIkCojShKeHD6npwbsDXfI0j47m4JHz5+RT3O8c+xsb6N11OQ5shwfHVIXFU+fPObu6C6T0Zhf+Pmf5+T0mKLK+aOf/jFCStIsZTwZhygtFKvktQB3GWt8P6aYjlBK0uuljCYT6rpEa4V1zVZvccx4PGF8cEyiY/o3bpD2e1TCYV2FqS2uKIKlbyzSy7C4yQHSoyOFqWuch7gXoyPJZJIznUxDpEUUnJROSyoMsTG0y6WtMTx88IC9mzfpbQxxWhAN+4zKgsFG0miuwV/Q62XIaUWvlxLFCaV3lLGnyEd4QhSHqWv6vZC2uK5qvHNESNJ+j7KAWAe/QF2FxRK6yS0znU6ItEKnCSJ2eG9w0uGFI8tSrA+5cxAClQbKIYlSyjI/p2znRva08BwdH7I5DKGSJ6cFP/rZezhXs7O1QV1V3Ni9QVkW7N7YYe/GbkhIpSQiihDOIZ1DR9EsKZjwgiSOkQqmkwnQaJ0Gjs8mnJ6ekCSae3duXK4QMAvvEAIlAz0gVIiZ77LOs3hoAg8pY42WmsFmyt04wyI5PDrh8weP2N7eYXNrQOp92EREKAQGJQROSpCKqKExvHBYAUIrNJIkjjk+PqXIQzv2ej22tsLEVRQFxlh8JvFSY4XH1QZpLKpxmEqlcK7pS97hpMclmqyfkFcVwqqZtdTr92cashCSHIvSFrwCF1IRCBVyy8xWNM+aMGjNo0mJ94IoSgN14yqEgDTJQFgWtbnH8/57P+Xw5IzBYMDu9g3u3rkLHZtvFnFiHePpBC8gSVI8noePH7E73GNv+wZaq3NFwntG0wk/eu8nOGE5Pj7g+PiI7a0doiieafpCNJ59B7mZ8PTgAOs9WZrRS3t4CzqOkVrhhGCcT8LCRdEqA1wC9i7AtwEBUZowmUzwVUGURFjnGE3HCAkjMWZve5+7d+5w+PwJE2NRUUwvy8j6GUVeUlU1sVTYvGI8LWeUldIK3eshpSSvCp4cPyWNYrTQPHj6hNqasPWjFBgb8uUroYiTiNPRSdg7ty74+NOPcaVhe2uLg4PnGFNjXFBEyzyntjW2jhD1iiyfvCbgjvOU45xpUVBUOVtb2yRpOuvgSZqGtLxeIPoxsp+QRj2MdxjnOJmekZsKhSBSYXm5R2JdyMWtpUJIT20dqhIoC95X5NJRlRWRitBekPXCFn9+UjI+OMHIgrTXC8mMlGJnb4dplZPJAQgYbm00eeFt0N6lQhNoiUlZ4Oo65E/JEmogUprpdEI+mRKrCNXT2NLiKxe0rERTm4q6LnFek8RxWC6OYDoOy/uV1kgVHDEWi1AEB6dSVMYyGp3hpCOJY7yFaJDircUbGygG33LcLc8doqTjKPDlzoZQwMFgyHR6QpJEGFORlxNqE6J3GuMZ8CE6Q6rGhxHx6MlTBJIvvfUWsfJ4XzdUR1hxY52nNBXIEKE0n+FRiPO6hbj/hrH1jrp2TWKp5uIZz+yQwhNHMf2teyGffLPQy0uI0wF7xjKZTDg9OWV4byPs1Stl4zwPXLW1lroyxELipUMhsJXhJJ8yno5BwObmJlVVU5Ulk3FYjKJ0FCwy58EYpHNIF4Z8XYVUGmk2oCprrLENzQUeST4pgyWpPCfHx3ggyzJ0FBFFwWpyUkAk0S7C+wglQpZHpJ05sJkRX8GBfXIyBeUZDPsMspRqWgenuTVEuqHfZrZW0/YI7t1/i7v3JVUdksC5YsSTx59RlRIvJXs3b7Kxucnx8RFHhwfYOmzlZ4zBVZ4nT58w7A8YxMNmy8Lg1/r80QOePn2GcZYsC5NXJGUn6iPM2s45xpMJnzz8hJPJCB1HVHnO0dExcZQwHG42zn/Lo0eP2d3c5e5eoIJCPzl/m3NcD3Hmpa2xAqbjKXGSoKKwcbk1lkFvwCSfMM1zTs6O+MmHFWf5FFNbpLQkw4TKhWCH09EZPs9D9JwUaCXZ2NwMi8bGY2JTE6mYjWTAwycPqGuL0pppPiWwFKGeOgoZj2pTY6wlI0Rmee+wwvH+px9gsSBDmGqImKpx1mCsx5Yrk+2+JuAuBLkNnvmdYX+2zZ4xIYujjhzWhRziWT8l2t3BjktKUzE9LamtJUsDZ2lq02RqFCBk0GN8CCiSQqJkBM5QTAumdU5VBHNqs9enjirKaYEqHcP+ECkk1ttZiNdgY8hmrBCECAQhJToKKWGtdZTGhA6DoLe1yeToGHtyQvG8Jt4ZoCNFRkyaKqSX1CcFljIAWqQDwRkp0iQsZJFKMTk54+zsjFhHswghZx1xKkFodKIRLix6MsYikMQITFFT+4opUz579jNiX9P6sM4l9H6tA3fsG54Yqen1+jg3IYo1qQubl2itZgt/WnBvS5Q+rBD9/OFD8umU27fuEPcUwjsQjiSJqOo6OC+VIm4W/yxy8M11DkyTs8XUJbWpUXF0/gZtpEf43OgoUGEXVvL54AQcDofc2N0OHHmUkzfvEBzsvqEx/KxQLwRWC0Qk6Q/DngF1WSFFsybAutDvZJi4qobHdbUP4Y4ONoZ9ahsjlaAspuQThbcGsTUIG5BPS4qiwmnL1uYmXoTEeN578jwHAk2TIEO4pm9DEWde0otkc+OI816EaIyGb5cqIk4D9RBnEXEcoaSa7RnbH1qsD6GrvaTPo0cP+P4//af8r/6n38A9/wH/3/dGPJjUfOe7v8wv/uIvkqYJd2/fIU3TWY4dpTS3928RJ9FszFkb6CYlNW/f/zKlqZFSYiqLNY6QTz+MrzDNOvA1xXTCeDSisBUYw/bGJlEco4TAWEh00PgrawJQqqaM2crZ1sRrYtwRjCYTDo4P8d4zLQoGekCbaz/Pc5IowVrLyfERJ+YwhP06UNaQj0pOxyecjsbkRclASuqqolSCqN+7kNakMjXgmU4mnNoRpjJEKqKsCkbjMUIKoiQhSWJMXVOVNd5BP8ugmYS89EyrHKEDtWzKEqk0WifBke8cSv6bsIhJQNpLGU3GnJycEMchRW2cxERxcAYKJxEuODi8DDzmje1ddJowrsJip/bDOn++mbFENbxeCDkypkJpRRJlIa2BylBCMxmPkE7Q39jkYHoAxrHZH87GkBQCLQzCWoQfNeAmA8fm+mEnHuepR4JxUZOkGZHW7G3vcnh4jJ1WyDiiLyS1CcAglCZsXBc+YH42QgjYvLmHNwaTF/R0TLKzi2xi0B0OISW2MhhvEELhbeAqkzgO+nFVgZS4wjEuxoxPRtzd26EnF3N0Ukp0myueoCX3sj61iZvkbcEaMuY8lrmlTZyz+GaRmfSeNI7DTkndpd0+UFHBeRrOaa0vbOgwi+7orKBtFa+wqArqylCVhrQD7mKGcQGURWfj4g68N9RFs9LWtAuqPKpZ6QugY0miBVo5rBCUzoCUlEXF0fFJWLJOoJxajng8HjWamSTXkvTObZSWTeii4O237vP2l8ImHcYGCqF2NU6HBGJaZWglOD07wnuPMQEMe73z/W77vR6ql6IzgapAVSEnTRzHDIYbJEmGVhqtI+IohFHe8xFRFjY1T3RQSKSOEFKdJ6qynWyLAmo8CI10oHXCN3/tkD/zVhooq5Hg//KhIssyTk9PQFg2BhtsDDbOE9wBCoGrKvLKzOYcKST37t7j7u27s1z2AHleIUTdOPBdmFiFxznD/f17DJIh/eGwySdEs57C43BUVUVZ28Dvd3K6z3wCvu0DnslkgvOOg6MDKlMz3NikrCtGZyN6vR5plgb/TlUTJ0mwgRTEOkbHisJUlGclkYjY39tHxpLi6BhvPSdlEXZuE6KtRnA8a0VRFcHHhCevppRViTFmprHjQtI+azylKZnm07DvRCyxdR0W0tkKbz2RB2E81BXeOZyznbG4WF4LcHfWIXyIK9/Y2EAIxdHhEd6DQGKqGh1rbG0xzeKZG1s7bPQG1LYm9pIk6wceazrF2PPBGzDGh0RPXiIUWBPCibABdIoyJ9aaRx9+zo29ClMVaB3hMhiPx3jnyNIMU4Som7qWlGVJ3IvJsjgMVK+QwhNhsNIQx4IqtxyfHLG9s4P1YWeoOIoZTUMoZmELahHKjrKMiJR8PGacT0iFxJU1sVd4KcmLgqiXkOgIV1RUeKSWCA0OEXLyVAZXW4TxwXpxHougPxyQpCl+SeiUEGEtgBSiYVcdWdYjL2OEELPtDK21zSq6ILPIDKlxHoRUZGkv7PkpASxtaoTzRU6+yeanUWrR5gbnJbeSJDHf+MbP0UtjBoMeM9hvZoDzMG4P3s7YGmtqsAaTTwEF6JBBkhYoHMKBlAYvmvBY4TDOgpfnjq7akCQx1jnSJEE2E2EbPWJrGxzk3mCrGuUaM0JKvLVgQqs2W28EqqUokAqEcZhySq/fD0nu6jpEUkhJHIf2n47HaOdQ8SZlZckLQ5IqbuztcXM3rGAtyoq6qjHWYGtPZSacjJs8PM4xOjujrOowOcmg/NRlsFq1UiGtcl1jvUf6kEfnP7iT4X2BKEGUBcIofvCDH/BHP7BYW6GkJJKSKI5nkWMq0vT6PXq9Ad47TF2jlEarKMQiCcGgPyBN06YNJXGckCRBa67rCilDRtRhkpJIjY7TEKLZbgYiXJNM7byP1HV97hD34nwicJ6qsggpuH/vS2xublPVFePphLwoZv1xKx1gnKF2lrIMm7hQeeqyRjpPz8egYqSKUFpgpSLOUrKdbUajU44Oj9i0m/R7PdI4wRuPmVahr0jw1oZoPa1IG8rZGRuwyoUd6KI4bP9IXUNZ4rwnVRGJilGRpigKJsUUU4dJ6HzTksXyWoC7khIhg4OoLCuqsibLMkxlKEfTsBNOYdBSo61HiwhnLYeHBxgbNg6OVFhUkkmNVh6vBJUxWBNC5qQSYD3eGYTWs5WVQoQMfcZY9m/exhhHlvZD/LoXIZ65sfCqusJag9aKsqrIhhtYqwPV4yHNEuJYk1RTKluSJppiMmWSh82ZN4Z9tJJMpiM8jlRpnK2ppgUJio3BgDhJmBY1NtVoAZlswjmVA0psYWfbbzklcTribDrFexikGdYYtAsx3QLbsOsSrWK8CYta2t192iBAJRVRkuCdR9SgI42KBuzK20CJ1mHlZZ5PQ/oBQAjFYGOTNB0AstmjQvK1r/ep65I4HeCEBa/xooYmKkVIRRTH4OVsJSbAuVewiW6aLXbx7O1usLe72RyvZyk1wmBuMl02b5qPT8Nm2d7grcWYivFoBEiUTtna6qE2N8L2imWF1h7rmpTIgrAwyytMo/1nWQ8rS0aTKTqKyHq92U5BSusQxdQkGqtN2Ey8FgJtHbKTyhnO86f7QJrjASMc1ntsVQUAaHYcwodc8d57hoMhUZoiQigXZTGmKhWffPwZn334Kbaug+Zrw0YXCJhMS/KqItaa3Z0tTB2sSWtLkrih87KoiSYLG4nEscRYUDrj6OSYf/hJha8Vtpjyu088tVBEePb2b1BUOaPxiIm3uGqM8w6JIFExdWE5HZ+gZcjqWjZbWoKgqkLkWaRDrLtWEXGcEMcxxhjqukJIiBqfQxwlDDaGs3BmKUKc+GA4oJ/2iHWEcY2/Kw50kxRBUVFKoYRg0MtoVz4P0luB/PFN7L6HsFdps/q3ob3C9psmBD+Y8G2rqiIvSipTooZh5fFonJOgcb5CTQpM4TGycTrXliwdoHQETrA5GCKUpK4NxtRMJwVJktDrZbPNaHyzu1TcG6CkZJBl1NMSgaA0FVGUMhhs4pzj9PR0Ja6us4dqCvwTIGmu/3967/9TIcQO8HeBd4BPgL/ovT9u7vnrwF8muNn/mvf+71/xEIy3eOMoiwAmSZKSaE95dIYuTciAl2ksDiksp9OcyXhEP8nopRmmLoniGF/WRITBUxiDtR6pddjAwXm0DPuBlrYCB0qE5FTChyW9xtpA/YSQGwa9YLLlec7B0QGbm5s4YLC5SZplPH9+gJKaLMsY9nucnozIej02siHTfEraj6hdhSnGVPWISFuUykkyj5CWuCjBK7Qc4dwGRT0AIbFAmnmUf4JyNcYSQqcSg3caawXeR7jKkEaaSMd4VyKlw6Eb52tEplPevvdlBpGmLgrKsgipFLxFNlsOCa8QQmEr0yyuKDCupqottZk0CceCRtuanl5Ikt4WSieN5iyxDvobSZh4cFg83qdYkTex067RaBuXbGs+dyj3872LQmx94GDDxBBi8F2goi5FQ0icdYxOn/D8+QEHBwdNFIshS1Nu7t/kxu5N8AbnHFWdhwlAtI7IsAS0RlIKTyE81lmsFEymU5QOdMdkOiXWOuzEVBQB3PGzfXhV1sNJRZ2q4FitDbImOLMlYfaQYUONojY8PTmaLTlvaUUpJarxGzjnmEwmKClJ+xFSeXa3MuLYIWVMqhRV46ijWfTlPWwNB5iGmkyShEp6kkxTFSVps0K4GXpAWH7gvSdN+wg9YDQZ8S++98f86NMht2/tE0UJngoda/qbPUTp8VHIpJ7nOXUd0mQnsSbWGt2swkyLXnC2Nl9wPJowGo0YZD2GSYa3sLW1zXQyRakQpjmZjinKknE+wU3GPDt4jjE2WFTQpOkQYSwrjbW+mSQ0aZbSS1PSJGFzcyNQvJFGNjRgFEWN5hz+Dv65sJ2kkCGcue1Pbf8K/1onP7PQVkeYUIuioiqnVGVBXcPJ6RHGVFRV8JkURRloYSmoq5qyKLDOobUKARb5lCSJ0VISR82COAQ9pZmMp0xHE+oyWHS3b9+mrmuePXs2y6G0TNbR3Evgf+S9HwshIuCfCSH+G+B/AfyO9/5vCiF+C/gt4DeFEN8E/hLwLcIG2f9ACPE1v2IfVeeCmSW0DOaaDOAapzFy2Kd6fkqsJXJaobQny/r0BxsM4owojknjGG8N/f6Ak5OT4IjygftyTY53UIF60BFeCtI4QytFrBRaKrTWxEmCcYZpmTMajYHgJJQIoiiYyKPRiP39W6RphlYRmxtbzSrHsNp0uLHBZ5894O233+LmzX2quuLZ06dMJzlRpBFSkmUbSCtxNifSU4Ssw4pJWVFWJUrGWCuoakskEyQKrTxSREg0XpYoXQU2T5yEpF0+AqmolcDLmCgJoZS9eMCjzz8hiVKGg02cVyT9HdIsJY0zkjgJ3nhfYGpHXVQkg7CEeporiipEb1RVhVI1Ump8E/kCEu9l49Nrd3U/j3YB12aZbYAnnJFeNNEqFzXbVi4sv55d4umuZm3v614bnPAhFPWdd96e0RoCGgdugtbBiRtphY8iQspfFRaJCUEhJFXjvNdKMKoNaZoyjGJUFOFMcO6WZcG0KMjSFJ0kRJGmrmsm0wlOKTbpoZrVryIOCbyEDZSZFWHloxCCre1tXLMiVWtNpPXsvZIkCfmFygqcpcjHREKTJs3Wee581XLQozw0Dm9vCxQhlYKti0BXNZt2n+/kGyaC1gHtfdi3WCVDdnZuEquYcpLz6MEjNrZ20GmGd47R0TG1qUAK8rqc1X0ynmCjmGhjMyzQGoUFiVIpjDUURYmXgq0bN4hU2PFMGMvJ6QF1ZRj0hgw3Ntje2eBsNGLSpPrNJwU6joh6abDUQ8p88nwaUmsbqMsa72E0PuPpk0c4a2eb4TgX/DBJGuLkB4MBm5ubYdKrqia/f0SkNb3BsJkEYnr9XrPBiUTKdvW1DFts0mjZUpBGPRhudRDtS1jrmoCQkIm2qksmxRTrLUVRUJQ5xtRMJuOQZ0Yq6rIiURmxcJT5lLOzYxTBwf78+XPwnuHGkLPTM/r9PpPxeBmkAuvtoeqBtpSo+eeBvwD8+eb43wb+EfCbzfG/470vgY+FEB8AvwL8yxUPIYljamtw3lKXZRMWlaK9I4o0tqo5Oz6l1+szOh2R5zn9wQAP9HpZyLN+Ujb5H4JTIrWCqMkcJ7xAySbRlbGksSZWgdfUWjGeTMmLnMHGBhvDLdI0pDsQwLA/YDKdst+YbGVZ0esNcM4TRSG3RK8XIlyMt9y8dZuPPvmUO7dvkSUJWxtb2NpwcHCIVIonxTOSJOHe/XvIJOSS0MJjhCTNSqaTCYYKSUwUDZGuBGGw3mNqhXcxogVH74ITEUB4Iu8RokY2i4i8LBluJkjjODr8lMk0x5pglkY67GgkEEgV0xsMiJMeXil6mxm7+28x6H29CVk8zy2STycU+Zg4ifCOsFagmAagdSEuWSrZUOKu4d1FoyFBP4kQSgctvFk+30ZtzJj8Wbw8tHAk2+iHWaRO+Ls7MSgVnH7d+GYpQ/pepVXIW4+7EFudVybwoaTUtUEIGzTsRBL3UqIso6pDSuMsyzg6PMY5y/jsDGcMfSHwSqFkyF7pXY2RoGqHroOSopM0xFYDJhbYKHDG5vSMPM/x3odoEKUaijJo93Eacn1HKkR+OREmU1kbpGlWDZ8nnrnw3t00YZ7A+84iTefmz3OuK2zasrW5yc7WVkj1oaAoDXfuvx24dSVROub5wXPOxuMZgPbSDOlD+g9jQ3KzfJxjBRydnlBVJVESM0w0QmuMtxTTEZQVSmimkzHGlOhYN68VLA8URGkU+Pk0pqwK6qokyRLiKEZ4wXBzSBLHuKLk9Og0RD+1lhCC6WRCVVacFSWHz59jTODD8R5jHc41/qM4rDRN0zhEKSVJ43eKm+NpmOwHfZIoCvsX0DpT23QWYW2GjBQ0eXGc67G5sY2H2V4BnuBvbDchMnVJVRVMpyNOTk4Yj0YoIXj+7FnYXyGOKUpDnPWZVhVRf7AUUmFNzl2EreT/AHgX+D97739PCLHvvX/cdKjHQoibzeV3gd/t3P6gOTZf5q8Dvw4hUsZZ1/yzzUKmHpHSOF9AYnEuZ2NHIMUEpRVbeylgcDikqHEqZG+MBwKlIiI1JLIRk3HVhOAFx9/NvV2qosSYmiTRRHFYnaq0YnRyyqPHj7mxv8/e3g1u7e9jjaHIi5BydjhkNBrx6PEThAAdRSjpSJKUo6MjsiRjY6MXnHYOfvTDHzGdTBj0e+zf2idLU5SO2N3d5fHjxzx68pjNvW3iJCHRCuegrCp6vR460UjnkVpRTR3FtEbp4BA0FUS6h1IRUqgm3DOMTyEkkY6ROiSeSkTG3u7bFJNjbFkQixDTIEXYnESK4Fy0ForKcHr2lKI2WOtCTLsLucGFEKRZSr8/IE5Ttja3iSJNL+uTF1M++PBnPHz4EC0k/86f+3NEkceaEuMczpkZZYAwKH0ORk46ZKP9n+/403ChhNRhWkYYe24dcAHkZ3dA5++LmfUUTm5ibYi7DqzdeVyolDFC1iHXlpCIVGKlQwtLLBVHp2c8ffZ8RvGcno6I4whjDNPplPFkErYm7PXYHA6JowiroNQeqzymqqGoUUKACsBuBVR5gXFhw+0ojtENTVDlBZPphDwv2N3dCdZGHJGgmuydTSRSJKF24DouaN+2R/NyzQ+lJN5ZBMEfc75gv9lztm0rZ8GHxXvb2zt8/PEnOO/Zu7mHIGwG44xlnI+QUrJ340aIDMmLmQYvZNgQxzYpQBSQxDFJHCOkIB9PKWXOxnDI5t4O05MzitMJaayoTcW0DE5D3WyOUrqwclw6j81LtBJIHdKPeAzewKg8I48U3lh6WR8QVGUZrHVkyB9V5SAhn0xQIqwlcN6HBVlnI6TSqGajlOPjI05OjhEIjDVhkZoXM3Dv93vc3N1lOAwL2WpjEI1iEcdJWKzUAH2/P5hZEVKIwME3UUpeAjGBynQJMMSzx1tvhzFgrcVUFbY2OGspyuDDaBc3/rd//x8uQOwga4F7Q6n8ohBiC/j/CCG+veLyRUHLl9y63vvfBn4bYLg99NPRlNIUITRLCSajMyKp+P9T92c/lqdpfh/2ebffepZYMiKX2qu6eqnu6Z7hLBzuI5GmyBFl2ZJpyIBhASKsGwE2YBiW9AcIoCFA8IUBA7qTYAg270gIkGWQEHdxJM6QMz29VXfXmlWVS6xn+23v4ovnPSeyarqbI4g0Wr9CVmZGRpw48Vue93m/z3dpSkN9VKDnVrqtIHAAFIxjoN91oAzOldiiwJSCI58uThk6j9FizqO1cGG1Aq0Knnx2w+3qmkePHuUhoZg57bodn33yMevba46OljjnsLZgc3XNYrHk/N452lhWtytilI5rsVhgtWLWVHSbnTBt/AQp8faX32K9XvPBhx/Q9x2LoyVN2zJbtlR1TVEXBANTDpNuGvGjCF5gAm0q2qYCv2LT7ShcgTORwhRY6/DTxGazFuaCsZRVSek0VSVdICmgYshXIMMEJDkPCMUTrVFWUSnxyWir4u6i7a9clGLvd9fsbiaefvw+kw84V2Gs5elnn/Hxhx9x794xdVuzPD6hrGqM1vhxoN9u6XcbJFxJDMp0GjPVRef3dWcwhpKHOWnNFDwhCnQgqMrdMPhw033uzf7BWzIky/X1NWcvPRRM80C1TKC0AD5G3EkxXqiGEbabHavrG0hgjcX7QFmW2TfeZoqol628scRJvIl0ECbOmBJXN7c8+eQZD86OePDwHiQt3B2jOVrMSUA/yvDUKs3tdkvfdWw3G1RK3Ds7E18aLTYVKkH0YvNQaZOnG3c/q9aGdtay2W6J8QW4RjCYjLPnxVC2d9J9psDV5VMimqQ0zy+e8cabb3Fy75ikEl3fkwjcbNZsOlF0juOQm46I0Yp+2KE2iaZpxc56GPBBHDCLsiBGGd4qZRmHidI4nNJQFFSNsGKm7ZoYE/0wYLVFh4E0FPh+ou86ykp8/00IhH4ioVFMhGFH4Sq63ZrgBTqbzWYSUmI0Zu4IGtxUyL0VwbgCW0hEY1UV1HUrMOO9I6ZJzOq894x+IkQ5f1OYuN1ecXH1DJ0cOuWsWGMylm9pmpq6Fi3H6ck96rqhqkpJYHJO5orWHeYAzlmM+kJmrNIY6yisQ2WW15y7z/mpFt75+B/Elkkp3Sil/jbwF4CnSqmHuWt/CDzLn/YYeOWFL3sZ+PRnvrDKDIKosMYd+KrVrKUfRkwsUF3AJhmG+inS7Xq2uzV1VYvyS3ecPXrAiDjkddFjnMNhiCEwDH0u7sLFPr13QoqBaZooi5q2bnG2YLY84uPHn/L4k8d89tkTHjx4QFVVpJR4//0PePjwIecPHjAMI8+fiTS4KArunZ7irLAP+r5jvmgwFh5/9BHtbMZXv/oV8cjWSpgDSoHR9NNEaas8/JLYwBgTZVlijGE79AzXa7rVGtNWlM7SNiXLtqUsC7bbDTDirKUua+bzOc+fXTCOI6enpxRFyTCsmXzHAaf+wo0xBU03Ttk6wPDk6TXduJUdVFUTYqBpakII1GXDUeNQ0QvMUlWs1huOjpbcPz+naRu+853vys9QWIqiom1m3L//gKPjB1hnKMuC5D3T2DEOW6Z+K57oKhGVEdw+CaSjUKzWO37v298m+siv/7FfpSiyhGrP0U4KpS34O69xpe6oc3tY5242sIeA8gxAK1KygEX7gLMKHcXqddVtcXVFyp2oUZpaG4kHzOHVe9+hhHTjNRLbrGLi8SefcXXT89JLL9HUmqQN+ESpNCYoVrsdT66vBPbJxnYxyM+x1xc0dS2K2whPP7vg+dML3n7zFawVu4XPu7SnnNO7EUZOXjSltVe0TfvCgPoPLoZ937EbRjbbjudXz3FlTe87Tk5OSCim4KlsyfJIFJn9IA1Zm2X1JmPpWIMrHUMMYomsYLsTjHy/QxmGgUKBJhK1Z9Nd53D2GqMjfgIfO8axQyVDXS8k9WrPdEOhk8IYxTDCNCVi0hAi0U+U1nK8XOIKyzp2bKce7aA8Mky3neD1GrSxlHVFSAPaKRSaaQzE5NHGknTCx4myslRNgbYNpS2oTMXp8T1IkU8++ZTNbuBmtabrep4+vcEH2Q2W1QdUVXVwTXXOiUahaYU6WdXMF3OauqashJAgtr4Gm+0ZnLECI7H3y/rZhR3+cGyZM2DKhb0G/hzwfwX+BvBvA381//7X85f8DeC/UEr9J8hA9W3gv/tZ30NrLdj4Pi5uioTo2W7Wsr2rDRWKuBvxIaFtISHUWrPtdxSlo5k3DDpC0gQf+fjpUx7/8GOsMnzzna9QNw3dbksKHiIsF0sp7GUh/GYMZWEYfGAxX3L/PLBer5nGiLWyLavrms1my2zdcTw/oS4aNus1KoFRmnbWclQc0/c9FxfPWd3ekrxnXlXM6obzB/fph4F4dc3tasXs5AhnhfvqY2Kz2eZtnXTO4ygD1vK4xRzV4pEeoesHjhcLnCuYtXMWszld12GVuMwtl0vZPmdq1dMnH2OsyHXEROruxogpsdt1XK9WHC2PmeLE5cWOq9UTQhg4Pjpmu91wdLxkGMRj5uWXH+ViqbBahslVUVIVMgAsioLXXn4Fpwb8NHFxveL733vCFMV+12hH07Q8fOlljo7mzE8fULiCfjew2a4JQWYnaEdSmpAUt6uNhApmXF6pFwzHVFaZvgjT7Av/5478eZlmmL9YhptRHqgcXMikIlFBPWuJw8iuW9H1PednZ5RlKapBP1FWFSjhWROF6il4sUJH6FdbfJfY3uzAGwpjmYaJZj4X/Pf2lsI4kVIZg0rZp0QbYWQkMck7M45xmLi9uOTJ48e8/vIDiqzGfWHqjAJCinz7u+/y6htvMG8bSmuZhp6iqOi6HVXpfkppUMxnc84fzNlud1zdrHh+eUk3bNjsNjx69BpFWdEuWrqhJ0zxsICO04h1DrQihERUicqUVJU4ZobspKmi8MqTInvjB3yY2A1b+s5TuBLvyTMIh7Ua00Y0AR/WJJswrsAYK9CNE1qyqjvKVvjpCUdVSszjbbrEUTFqw5QmSuWwixqtBmLakvRITBbJXddMdkAFUE6sESbfo2xiedLgjGFz2zEMO85PT1BV4GR5wtXlpxigriymWJI4zkEtGiZP9CK6urm+yRYEAreY7CqqrWW+WNC2DS4vXjEneTVNS+EKlvMFTdtK3QPm83kmB/z04w/TuT8E/rOMu2vgr6WU/kul1H8L/DWl1F8BPgL+cn6ovqOU+mvAdxFy87/3s5gy+WtQJvttJ6FuaVViCiVeEaMnoTBtRW1rTHJsNluscyyWJ7jSEQvNNkyorPKLQahY2/WWKePYr776GkZbVje3pKQ4Ol6KxbCrmKbAMEzcrm4ZhoGjoyOaZsZqdctHH36KsQKbXFxc8O677/L2l76MD55Z27Jar3n+/Dlf/sqXMdZwc3PDar1hNpszr2uOFguqzCMHRV3VtG1LTyCoSeh0Pgh+l0RRF0KgrEqMM3gFPkaMsfRdR6sKxnFi1s4oChkW+Skw5kT2uqkPmJ/Rhqqqhb7l/R847zp3uGVVSdhzku13VZZ4v4eJpHBYa3LKEuyVQ/uvf/7sgsePH1O3Je+++z0enp9RtxZnEst5xXLR4IPQM1frHevNLT/+8S0xRTFPsgWFqzi9d8r9h/c5PXuA1RKVWNYt9+4/ZLtaCXZCyoydF9kzcnw+tGHPBgEOpV/lrl7yKlOGgA4WvQ5GHfEqZPpnyhCcdNJ934vopKnlwczKxKIo8KNkqgZjSeQOXBucdRhVkcIEIRJ8lHwBBccnx2JdMY3yPQpRBe8pjCFkwds4sr5Zsbq9Zbvd/vS+Ldf5zXaXE7QClZPHXGsjLojpgNB/4UuFELrd9Fxd3dLUtTQoRE6/dJpfWyL4UkyZsiyxhj7HyPVb8TAvtM2MI+k2p10HeYaktZEiljFoXZb4MZGCUJeHTqGtwVU1UVn6XqHMhCkVySViqRinHh8CrvL4Qqyxo7ekkLAFTEY8P7EGVZf4YWIaJ4rCMQVYtA8p0Wxvt8SQKJqKTZCErsJa+k3P7fUGP40cHx8xbQKrzYrLy0va2Qx3brl3ckphLM+fPePpk2d4oD1agN4z88CgidGzWl9z72hGXMizb4ylGz3dOIm98W7D1dUFMQrjK0TROzhbHH6fzWZUdY02RpqMqvxZZfUPxZb5PeCXfsLHL4E/+1O+5j8C/qN/1mu/8BXi7maFZ2oy9zkQcEVB9GIm5G978AqrRPjgnBht1fOGnU4YU2KUJkxejKQqx7AxpBDotxvsg0eEZAlJ8/TJp6hnMoA8Ob3Hrut5+uwZwQe6oWc+n7NcLqnaik23Zeh7FrMGP4589NHHfO+7v4+2mnfe+RqzWcNu1/H0s09JaWJ5fMxy3tJ3HVXbcrteoa0mhMDTi2ucM9x/6b7caGMkqoC2it2mZ7vd4v0kftF+ZH6ywFiNsxbnCrbXK0JV8PTiGR998D5N03B+/z62KPDKgzFo69CuoOt22ARd79EqYvX+bOeymPnCzrS4QraBU4hc3VyzPBZPe61EIRpVEhGZMYfUqpQSBE8MnsXxMa8uLWocKZPACFoLbm6UeIaYBMYoitJxbGZUVcHxvCUlzzB4Lq83fPzRD/nxj74HKGbtkvsPHzBfHvHn/+JvogBjNWPf4VXBsFkRvDyQISGpPvqOsbDvalNeEIy1aMR0zRrLiCaR5xFJCpsxYkNrcKyGLcM4ibYhRKYQSCHQ7Tr85DFGlL32BezdTxMnZY2ysgieP3iIZkEMim54xpQ8lApfGuqiZbNZM/iRzXZL27ZiBWEtzplDZ7YvxsuzY15G4ZXKEKD+A7uTPdgSNTlnIJKUznklKhuWvTCAPRxCvRzGyDgFxqlHpcTXvvwVgkq4umAIHTNVopPi6kZyQK0zKBLb7Y6UUvauqWkKl4fa4oJZ1hVGa4EcYmK37USBWUlqmnMVxVLl8KiEKQymMCjjqFwlOyMDY+zxXqN0RVkkxnXH+rZjHCasLTlenpDGSLuYYVOgcI6d99zerAgh0HcDs6qhaEqOy5bH7z1m8hOP3niDcezQ2kEITOPA9eVzZs2M1fUtY98xDAP3z8/44Q9/yKsvPWQxW/DJpx/w0Ycfst7uKNsFZvS4UhZnm8TyxDhDWZds1rciWAoTPkyoWUuhHJXSLMISnQLkTObBj+y2Hd5Hog9MQ+Ti+gr/TMgIH3/8EW3b8LOOnwuFagyRME7ouqDvewojbMumqbFO+LOpC0Qf6TYdx8tTdt2OoiiYvGe73TIVoioMIZAmT1NXYkGrNU+ePGGxmPHGl8STIyZJr08hsDg9lZ5OCUShjOb4WChLygg+dnp2j+urK1KMnN075fLyitVqw2zecHNzyzvvfJ2PPvqQx588pq4cR8diTkUS9S1KMQwjVTPj5OSY5xfPUDmWT2nJuSSIw1/btjJ4ceJPMaRR3BinCTP0NPMWg6V2BfO6JiUYx0CKA1VZslgsiTHx4Ycf4pzj+FiwUucKUhgP5zw3YRm+yIlJmYHS9x3VJHYQ2hisjENIPvtZpGyvijjvKa149U+8yl/+37/Fj37nXf7R/+P3Eez3jlMdfODFcqK1yoU0G7A1FVq7g5owxkRShsvLp3zvB9+nHz3n52fcv/+As7P7PHrlS+jk8VPPNHQMXcfoI6617H0SVWb59P1AUorFcpFZRXtqpgKVBHPPE9aYIkElJiJYw3KxIKYkQhTvmcaRvt8Aibqeix2DzlTApsFaQ0wKkyJmCoyrHUE5oe3NG+yipshy/2HXSdOSIT9rrTBMQqDv5VzvLQ5SlqKfnp4SfMLUxed1Ai9WaqUIMQgl0eZ/SHsG0k8eOyulGMaREBTaWGEDpchrb71BcdwQbcLEhPGJq5UkEYXg8WHMi6VQ/sqqZMz+Jz4lbm9vuV3dUlpRnAYfss2Al2dkmFD9KDbH1uYUMoUqNLt+iysVhSnZbWX4bkpDysSA3c2aTz56zLgbOT055uzhEaUzAtukwDQNTGNiM+4y5TkrikNiJHK9XfH4s0+YL5b008h2s8XVNdpHbi6eU5YF987uEULg1k+cnpwAIvKbxomPP37Ms+fP8EGRsFze3FDFwKOXHmWtjpxL55RAKkl2yLfbDbapUH4Ea5i8J06esOmZ1y3LZkZwkemmY9nMsE4SmqqqxvvI5CfGcWSz2f7MuvpzUdxJYG3B2E/0mx3Pb9ciO543tGVL8BP1UYNyhqQU9WzOadMwXxxzcfuEbd+jkEBhlRI+RqZpYr5YMGwGdn0v0/7dlmZ2LCrMFLi5veLk0QMx88kXXmvN0dERPgSstcxmM3F7mya2V9e4suDN117hyZMLnl9f8eTJM1566RFt23BzfU1Vtey2HSFuqcocDKE1226HLUumELndrnl+dYGqC/FDD4K3Nm3D0PUEH2hnM0yhUV7RVBXr3QY/DpRVg+8nugSh2/Lmq2+yuh1Yb66Zxp6joyV9L0W86zqWy3h4EH+izVCm5xksKqlDdzxNAVdUaGcwyhJTwjNird0T6FDaYkyBUYHZq3Pas5Gz186ZnSzQxpGIBIThIYPMkDnDGlKJVgbxfEkIUXjvCSSLYjufc3J8TAqO73zvB1xfXDF1PR++9574lRSOs/MzHj16xPH9l7HWEYL42Y/9iA9e7FVVHiDmbj4lYZpEgCTahBRHfDKMJGyMOGAInmdXlzhXHCwGxkFEO4KJQtu2smNQcHX1hLIsOT87h6IEFzl+6YwYxXIhGk1A6KU+JXo/0TSzQwf+ohFU3+9Tdu6sfF22ca7b6m4onl4YkL/AtJjGkfV2DWrOoo2EKLjvT30Ec/GPefejraXvB/wUqA5bgcTASD2vKWfSTW/Wa8Zx4vj4GJN3Gq4sGaaRoRejrHk7E2OuELPStxCFtlKMmx1ljLSF0HcDmt0ozJRd39MqxWTEDTbgmVKk7waapubxBx/w8eNP+IV3vsbp8RGXl8+AxKOXHrK63VDMa+yswnSBuNux63uKwlG6gt2wY0yKjomHp8dMiK23qFYT1azBKYO2BuMsDx494sH9cz756CMA1ruO0T9lvVnjY8CWBW+89gq2LKTpyY1PROw8rLM0sxm3N7eMKTCbzdhFT8w2B7UrKWcaP4z0vZitnZ0cs7q9YX19xTQFzs8fcny8xC5qoRifHfP3/v5PL6s/F8VdW0MycvO52nJaHh2w3GgTVVOTtGI+a1DLI2rVYtCcnJ1z1V9hc1SMy1ila1uIAeVg0+2YF0JP2mw2lNWCGBPGWMZJVsCiEBKeMfIQHh0dMWXcq8gY6NnZGcPtmpQii+WCWXvE7ju/z831Ld/97nd55513aJqWxUIGjz6nN81mc8qqZLfzolA18Ojll6AymNLm9yIWx5tN7kDHkX7oWZ4uheNnDDoYzpYPaKqaJ8MTEolq1hIyvvrbv/2POT+/x/37DwB52G9ubjg/vy/QiytEqZiPF7s3wfQsIQuf6qoiEXBFKZa/VQVAv+vEsAgFStPMF+jCocfED/8/H/LZx5/SXyTePv0G1pb45BHS38DekkwrYZpozQEmkjchYMFhRpppegp49aVjXn74RwVacqJ2XN2u2GxHnj7+kI/e+yHGWk5Oznj1jbdYLJac3JMQ413XEVJkGoY7rDndNbqaiIqK5A1GGawfUWVkCB6fIovlEqONcIun6dB5WmvEbTBGqqbBWsvR0bEwQaaISZFyjBRag00oMxCBEOUe7cMk3GVkuAa8ENAhO4H9UM0agzWK1WbN0I88/vgT/sjXvynXMd2Nkfe7MaXgzddfp66FocF+/kD63AJwd+73M4dEVZRo02CtQ+mCdj5n6HYkY3GlMDbGYchdfsjUPmkklBIr2wSUZcliPifN2gzDiIhIGCCSr1uWBSoPrG83a8ZuxE8J1VTUxZy2bSSqLmNJw06iKWMUTcyYAt/8xjd4dP+czXqNVoqLiyu63Y5u6Hn0pVewfsYwjVhnmc1nOCe+VHH0DF3Pyf0zqlaGpe1shi4KKCJtUTJmP6GyLKkLmS3sd1IffPAB88WcdtZinBV/nKoSoaEXa2ZxP9WMQa6jD56b9S2bscc7Q72YMUWJ2auamuF2w2q1Yj6bk2JiMZ/le0HCepbLpTDsxhESLBbLn1lXfy6Ku3iFO06OZMJtrT4oCCOAl1Brny9K2y4lYEpJiHbyiXEYUdZI8o3RWG0JReT8/jF68DRVg9VFDjnQVE1DUdVcX9/w6OWX6XY7ylrUgLZwMvk/DOcUTTOjqGu63Zaq0ixPlzx89Ijv/uAHPHl2yWzxKcv5ks1ux/n5GYDMBcqSxdExyljWmx3OWFKhUVqx2e5wtsD7iJ88zhjq4yOiElWeVyPb7ZbZ0RzbKigSpS2xpcPGJDTPmEAZZk1DVVXc3N5wcnJOXTdoxK72/NErMHVsxu4Ooz3s0RPkeffeKXG1WjFMI/NZy+nJnNKJenjyYlCVVLbW1YZEyWJhadtj4hoo92E6CTDic3NQg2bpTGa2pKRQShSjUm0Dh/AOpeTfU8TZkHVLCfBYnSidwS5qlouGsnTEFLm+3fL73/5tttue+fyIV197jfPz+zw4P0cpxXa7pSgrEjBNgRgR/QOeqUyMccAbcRDUaLY3grkbbbi6upKGQ2naWlz9xnFkt93SDUNmtoqCtTp/ANYwKoExTBCGTjSKoCFkT9lZ0zDkhUKk7TnEOSa6rj/AMkopUsjMHK1RTtMNHdqWlK445NveFfnEmy+/zD4UJE5T1gFk8Zj6yQU+KdDOYkyJLUsetjMm3xP9SDICL0pgShL+dVFitZyHLguv5sbijBO4L88hpmkkKNClw2mDUhKk0223WKNxzhIrEQ+t+xWFT9RKqJXd0OOMWDuEKM6KVSM24N/6pW9RBQnLsFbx7NlTMYkzina55PT8HLRiNmvZbHfiHJl3O64Ql0ofPFjNLsguI4y98M6VpmwbdIDaORazlqosaZuSuirZrDdYpzHWMpu3zI+PiNmR3jihbKs83dZa42Y1T6+ueHZ7SVk3mELEdLO2lXuz24lttB9pkuett74MIVAWOVQmwrPnn4FRnN17QFNbeME++ScdPxfFHWTlH/v+4Lq453nv7UL7fkdhHcdHC7SxdH1PjAltDP0wSFB20NSlcNJ9SIQpcHp8zMVnT7m+uRF2iDFUdU011hwfH3N7fcPy+Ejwb6XE3U7r7CEtvg4AhSt49NJL9LstyXtOj0/55jfnHJ+dEYHj5THOSUe6WCwIUVJ9XFmLAAXDNPSiNlOgjKIsGkwybNdr6c6cZgojYxgpW6mSZVvhg2fse3Q0nJSn1E1F6CcM2T7UinWptZZu1xGPIrPZnKPjY66urnn48hvEL4YWqzuJv9YxK/8NlXW887W3mUJgNitZHDXC9AmJqSgoCkPCy8IKuajIAFxzt73ngBLE7EwqgzRFwlmNsUjeJXeURqVkof/8sQ8G4TAFVAeAaS/mSDRlQdM0PMAyjBPX1ze8/6Pv8t3f/10WiyVnZw+4d++Mtm4PD3gMYgyGCpQpUZEIyjJOgc9uroW1lPH5sqpEOlcUpCBQV1EUe6J8prbJa15dXlLXNSd1A0bhjWEvu3IxYUOiSJqn2w0Te497eQaESeLo++FOym4M86Zmt92hneGdX3iHMipMkGSeF8v6Ye0m21IkSOztDMgmZT/hAdxfIyCqlJn6HN6Dz/oBgaZEcGXKkpBkJzP0PZvVGj8Fjo6OJOTGaIZJdqFaZ5+XlFDZ5yZmXH5/XrfrDe285eTePXbTeJczgMyixlGYPqMXa+NF1XDkahSRrhdqakqJB/cfcHR+yvL4mJv+ltVqzbNnz0kpMZ/PcdZKOluCdrHAFo5d3+XgEWn+GCRBbJEhJdKdS+deuOWcEb5/NknTVuNTInhxlUwp4ZzFk9Akrm6uKeuKV994jXaxAKPz/EEYc4WzNG3LZ0+f8dZbX2WzvTwkM93erqjrht4PPLj/CtfXlywXxz+zpv5cFPcYpFtUSfjGUx/YrTthPgDzmWK1ueSkXbLdRjbjmtB50umrTDFRtTPK6Jm8lxXVWuqmxmpL2k0olZjCSLdbMU0PCUHCnYuiZJcXCWMNs9mcyfu8sAhkEjNWKfiqeIlvVmtuVitG73n+9DM+ffKMN9/4MtfXl7hS8dJLD5jP55S2pFcabTQx33h1O6Mb1iznc0JITH1PVZVoLZfCOgVOUdYl2inCJPmqhSto20Yi8XqPNXN8v8Nqw/FyzqOXXub65go/RWJQ3H/wgKZtuV2tD9v+z0Exn/uzdHY6gVWat998kM0aE3u3RCxQFuSqKItDHGRw1Q+E6POr3mHE0oXrg+AJwJpEZTOrVqXDuzoAB3v1nVa5E8qUx2TuOs4X3vyd3a+IhlIYMClyPG+YtxWjn7i92fDxB+/yo3e/y2xxzC/8wtcpbCm7AjQhphyYnuj8yOVuw6ATddMwDENmsMiQ3xpDU+fEnWlkmjzT5Nmu16ToeXB+LrvG6NkMPa0rsUmGkfu37xWs/YDX5Og0ob+BGFSFIP7eew99BQzZWTB4T5+gMRWgspbhhcFq7tZT2jNjcoC2yT786vNXHpWyrYycP+VHVBpQ2hCJsvAVpbCiklh7FJVABcM0oRDH1Jgi7awV4ddqhe0MhZM5VlVUxIgU6txEhSC2t1qpu8GjLSick4xjV+QAnp66rA5CwpRE4BdjRKVEn1ktq9UtWmteeeUVTk+OmJ8sGOOUcfaCl19++bBQxhBkxxMTbTUjRbEY3u12dF2PsZZ7y2MKpRmHnnHXUxeOy4sb1psNu92Oqp5RVQ0fffgRD196RGEd22kQLQeIt1K+HtF7rtZrbm5Xkt/a9XjAh0jTNtS12K/crm5l52RtNh0bQQle75xAy0dtSztv+PCD79Nth59ZV38uijsk2raRzsZZlDKyqvusIK1qGl/jdWIdOnTSDH2kz4OXruvpdxuZpudszClOOGvRTUEzn9HFNTF5YgRjCo6Pz5jPj1gu7x0w9pgiu13P+++/T1lWHB+fcH19TQiyA0gKbm9vmMbuIBWetQ2FtUx+out7nl9e0/dblosjXn3pFfzgOTs7YzuJtecnzz5l60fcoqQsSqx2VE2N1U68o8PEvGrRToZvo4Fu3LJeramLhrmJPHl6wXwW+MZXv4JJhs3mmRSicWDoRq4uLjg7O+Pk9BRtDMOwwvHTpQbScXMo5DLC2OPfX4BxkEFYBMLYg7KEsRdLATJWrvZ/UnnIlv8tDzT3QqTD1f8cLy8dPraXy0OZGe05kOELqMK+uwx+ot/1eadgKMqC0joWD1vMK4YQFc3ylPP797m5ugWn2TAxqcgQc0pS8uiqYGGazBWPByOyvWJUOwNjoq3nDP1AMcl9urq5Yrvd8uj8gVj2KkilZUB2RipIAY1KYcqKxluij/TZOtg5UWeHELHW0jSNwBEh4PNC4QJ0OjFaRWEMU1Jor9BeIuFkR2awxhGCdLpqf35Tzl5Nd+daHRYGWZCHbotSvQyelSY5jW1K8IFxCnTjwDCJDUCKQdw2rT04cGpt80IV6XYSZCjwjHTqKUactYf7RSOzNW88CtisNyzLgm6cpKBVBUPfH8zVxmkijgPOOrZTR9eLv0/TNBy9dSyOsYuaySn6OBGIbNYbiqKUrt3JczaOoyykWmGdZdxuKcuS2UzsIAjSjqw3G0ptuby85OLiOZvNCmssfd8BipPTE+6dn7MHSGw2fkvpbscZQ6AfR4FpiyqrcOe5iMv7L4qCWd3w9OrTbGdhs1+Npq4rSJqmmbHtdjhnOTm5x27T/dRnGn5OintK0JQ1ppAtyDh4iuxVXFc1s6KlLCS2y1SO6BVJb/n0+ackIw50RekkjsoWlGUlWhciWkWMUZzcO6GqW7TRTEPPiIgxLi4u+PjjD7l3dsryaM6Tzz4BpXj5pZdJKbC6vRXTn2MlrAgQcZGWzMjlcs6rr7/CBx8+ZrfdMQw9F9Gjo6J+80sieCgs7bwl6MBxfcLLX3/I9vlTpoQsVGPPdtuJvXCCWWpwk0NrQ9OI4GF2MqOpZxAVaVZwsbvk3cfv8fW33pEhlXHUzYymUQzdwOXVJS+/+hpaGz768GNeeXj2088/BpQhpQmUwZMgm33tUZKUcRajsoYzpdzViq/1HZ+cw96/qcRxc7vb4b2oh1My+KDk2iCe3Cj/B+l5iTxI08LPJiIWw/tK9YUOVKaCeYegSATGoT8wZErdkJQoAZW2JK2IOuJToAsJHcX+WWuHzruQsRvYrDfsYwFTEvXt7mbNZruVkGWtef7sOVZrmrolodjtOk7bhUSjjYGpMAQtg3G0xvcybO92HZvNlr7vqKuasixomjo3OIJXp+xbYpxjigljCsoIRXKopAQLd4kYBIPVUayAXVmSur3NrwxStVKUTnxZMpc1w0UvFHitsM7JEDB/1KeAKixWK+rsJzP0HZP3LJYPZTCqclyjLWQuE8XvJiXpiiUQR8yv9guxZOkWmZqaoCxZ3nMorSlqCc7wwdPM5xgt+hWXWUPkQXLZNBirqF1B9AFrNVP0jB7Ww5aLywumccKYTsz96prV7YoUIqen9zhaHmGN5t75GdMgdEitNLoopLsmosLE1A08ff6MfruVnOTJs91umC3FEsEHuWe0kSwFFTkI6QKKYZDFsV44jk+OcWVBt+uk0LdCOQ7jxPLsHsbeEKLYlBdVS90saJpJrk9bME6Rsi5QX4Rav3D8XBR3rRTaOkKcDhDCkIc0VntMUXG1fs58NqMMgUKXHJ3OOF2c8OnlJzSzFp0ifvTs1p1Y8rbysKhCUTYVpXGUTcO27/joow959bVXODk5ZrO55eZ2R9O8hJ9ElXh7c8vRcskwjmx3G4p8w3a7TjjJVSX+z2VJ1++o53NuVisKZ9jtjAx/x4nNZsVs3jAMHSF5Lm+uSBrcZzWFaUgEAolhmkApjLOUjcMunBQ7lViPG3wIlKUMtLq+Qxeak6NTOj/y5PoZysq0XoLCa6ZqZBhHVrcrjLEM40A/DJhDE36XdJSSYrXest2NlG1B1w/897/7uzRFiVIwn8/o+p6mneGnieP5gkcPH6AYICqiEhaGRhGV4XbVE0Li9GSJdcJjBw7c823XsdrsGMeBtix46f755+v0C4eEVkd2u5GPP32MVoE333gthyFndenhk1X2juFQ4GUNikQUfYDtbstZlOGmyiZdRXL4cQAiYZScVlcorro1u75HaU3XCctov63fUwr7rsdPAi3E7JVe1Q3dMLJ1PXNXcP3shhBbgoooE1geL7jd3NJNvdg2G0NRVoQQZO405sD0jJMrpQ7GZNZYvv/hj3n62VP+5K/9Kk1V0laOGCahLyK7MJ8073/wAYvlktIVOC3wDgq6rqdu7ozhXpytKiVQwTgMrHcdu2Fgee9EgphTZJw8650sRuKr49juOoZRIKOYwzQknSoeYE2BtO6e9T2DJ0bZtVRlRV0UhKEnKU10YqGAEorw5D3JGBkOI1YPRsscw5tENW9ojeO7v/8djs7vcdzeE/WqMTx4cB+tJMN2GAaGoefs7OzgQDqO4rBotCH5hLXie+Mzk0kpxABt7KmbCqsVt+styhqKuuT47BQfJsYQZJ4QxPIj855ICZ4/f8677/6Isim5d3YPWzhhwFhhTnVdL5RpBUVbUwfPFAJF1bDtBpzT3Nw8hwjbYYD7BX3fsV3/T6Fzjwm/7qQ7i1Gc4boJq2Ws4yefeb+JKTVUVWLRzFjMZjzbOaaxp98O3FzfQFLUObbqLrJMYY3Yjbrs5rfbbambmiozZLquo21b2rbl8ePHXD5/jnMiqkp1y9j3TNPEMPQUThSYPgddj2PHvdNTVrbg+OSEoetZXV+LTWzwLGZz2uMF7XKBXUjgcIw+D2oUxmq0kZveFkbMkKyS7miCzWoLyzmzVrzSy1SirdA3Q4g5tEEecFFLKrp+4OOPP6Kqa5rlDF1YyWb8wqEUpCidsYoQvef6+hp7ckQKgdmsZpoGui57rviBvWnTYX+f4+SmceTH771L13X8yi//Ost5xd0UVAatKUZClDnIXZTeTzsiMQW6YeQHP/gRhVO8/tqrYF6EbOBQ0r/A9JPPyf/q4ebyBu+FETSNIzkNDasdqEA/ebrQMyWBUoq6ImUL6r1yVJwpTR6sjcLo6HvGYWC+WOCDp20avFaMGq52W64vb4kxcnZ/QbWo0YUV/UaM6BBw2VVRvsdeYStMmWnyhwxbrTUPXnkAVmFKi7ZaovyCLFZ7ykyKkR//+H2++rV3JEKxcOKmKv7Oss/5IrSV/5PCFuk3W548fU5Z1iyLI3z26tnnf4YsutNGAk9sVmB7H9jtOnbbHcvlUrrSzJrZ70wJcW/Of2DbWKWxVYEn0m2F1VU3jQjItuK5tBcl7rZbyqrKA2153vvrW3wK6MIRtMIhYidKYdr0Y48Pon2YzxY4LYlMWsmQeLftSCoJM2ecsrpb0W3WMHqmcYCkuL1dc7ta8+ilR5ycnFCX8j7CMOBDEHW8EarwXiF8fHrC17/xDkVVYpxmGkdsplb2Qy+Gat5TlyWls9j5AlsWXD5bE32gLEUDtFwsAcPyaMHYXzCb/XPwc/8XfSggdgOj94zjhHNFjtWa8ClQny4oqxplsym+VUSnCFoRdSKqyJQm5vMZ2mRrYGQ764qS4MTrQ2esN8TAe++9R0wRH2TQuV6vqapK/Buqio8/+oiXHj6iLEqcMQQvPHVtFH3XMW9bGcxqw6KRYr/dCMbYNC2Xzy947/0PuLq64Pz0Hg/Uy5iZ+EJMmXpotCV4n3FkBTHR7QbMIDOD7WbD2I10O4naKnXFfHksjJL9hJ5MlzMGpzWb7RZnLWUp8JSxlqKV8F0/vYh930EbWsvkX2lNJDGMHqMdzhY4W9DUDTGK/YBRWcmaB677PlkGeQnrDDPdSJreCzi6s06CMFDCJ3f6YJD2045EzCZnwvxQGARY3u8GyAU+wzdf8EzZsz1A1Mfj2KGULKTaaIKSncugE0HDVCsoHB7kQUwFKQasIUMzhuAFgxdceT/81RRVxTiJkrjvOl555RUmrXjw+sscPxi5vr7m5GyBqw1WzbId7ohSmiEMVHWNteawXu3TgPZwxr44Lo8WnJwcU+YM0iEGNBZiOCxWQM4nFcphIp/nDCvKCqj5/MIqi9cUQBtP1TQc3z+hnBdEF8WVMmi61ZCFaJJchjJi7mX0oVs3RsLsp9Gj5zq7oeoD5h5y/J/NDpIpSZi7dg4NuLLIanFDmiaBYV4Qee1DTRICBaWkuL6+ZT5fUGhDayq00ax3G8I0obXidrVDl5J/8PzZc/zoWSwWNE2FRhPGROmKw/Dc6AlnLMxn3D6/ZLvdsbm55fLZpfi2591WSEK4CCGyN6vTewp1nj0prZkvF7ITswplZEcSU5QowqrE2EL8q7TMLYwz9P3AsNvhq4pu29F3PSqHAz199pTjxU+HWuHnpLgnBaosKIwBpQm9sBiqosaUlqauiW5OWTiUFe9vnTRKJQyGOARhz2Q/k6Is0GZvEKWwhSMEcK7Eh8TlxQVDv2O7WqOtPQzMqlKUfw/u36ebzTk6EopkSopxGilKyQt1VSEXIemMGzpSEhhovdkwjSMJUdgZqzg/O2e32zGfN6Akt5UYGYdRtn8JxmFgebygnVeCdY8JpRrSvOVUn0JMFGUhRWHXMwKlK6GBYRyEehkGrDEZAxV+uXGWKX7e8fuLhzYGk4rDoDME8DmRXRmLKyomLxF7xgrWt+/8YoyHwZFKUJYuQ2spu7Pn7e8k1LIMzqI1hwf0px0x46ohSlJO8FFsAiBvOe6+Wuko8Wv5AdvH6+0/I0YZxkmQSkJpxRgDXkEy8hOMk2caRo6XS26njQS3G33oSodhpHCilQhBYMNpmmhy8dq7+U3TxNX1Nef3zkgpYmrHaXtOWRUCiXkRV91c3xwGxxJSY6ibipTI4RAyAN0X9q7rDn8+Xx5TuAK0QZWKFECHhAqy4PoYCTGhozhLxigJZOawGH/uCSSpmHc1Yvtgy5Lu8jl9P2FcgQ+BXd+x7cRDRmthGWlj6DpJFhrHkRgFw6/rRiyBu/5wTcqyZBgGQPD/7a4//PyFk6SlqqpEDTxN9NutsG2qKidoQVVWwpYJQRZTrelixFvNk8cXLObH6DFhjeLIzXh+ccGzm0uuhx0DIi5ES6jGNE1Mk2XWzmgqCe2Zhom4N4YzhqJy9H3H7fU1q82GwXvO7t+XwI75HGXE8RJFzqXdNxORFPf3Y37O9t1OEIirHyQ+sK5raUq5g0y11swWczarW1QvorC+H4hpgBQoi0pCvX/G8XNR3GVOn/DZDL8oZAVXSnBRazRFl7BKocaEn0bcWGCWit1Nh5ois7ohRKGn7UKgLMQzORoNRmV1mWMaoalKwrSjbWrOzh9ytbrg9vqaME20dc0rL70sW6Wy4PbmFqUM3TBgi4KirkF7bOnAJ7EMGIVnf3y84Ob2RmYFhaNMYjfw5PkzbF1wlMB6YQh4H1FJyeBYa9KspFiWxDjh/URIHnSScI5CFquyctigUD5RoCmSKCFHFUFDPXOUhSx21pVMPpGspqwKDoD7547MvtYiXpGLIQVqs+s4Pj0SFz+VMCrKNcm82xx8l8t3yNcRXOlQPvOlk3hjy2A05oGnODJaJ11MVOlzqUBSr3NKTZLPHyePD0Ip85NcW3n9PaMnM3TU/jVeYOIgy0yIgdELJzxlWmKZDHUM2GgISrG6vuX25hqHZt1tCFGoav0QIQrs9OnNZ1nU1jBmD/y2bQ4QYEqJ1Ajz6/b2lg8/+JDV7RptNL/0S9+imdWst1tM4WhnLYQ7e1elFLvdTqA2pdntugzNTDngwR4UksF7YlQ4pXJ8nmBSKaUDLr3reypV0NYSqxgnj3UcoKr9NVPqDtbSSoSAxCCwj9UkDTEhM6EkauU9PBNjZLsZWK3XrFcrnC04Pj6maqqDpcf+GIYB54TNFkLAOimoCp0zbo340iD6EqM0LkNTkw8EJVJ+o+5mEcK3LyhPjjgdPeVM6L9HiwVd13F7dYPThocn9whpxCZojo6YnOwkjBa7gaijONNajQpCJx2iDICbowX3YmSKgaPTe5wuT2jaBpcXep+bGSEiiXo25ZN8pyXJO6VIVth6Cis7CRUjKQ4MU8yUUMvkBxnYO8N2Lffi3uKhsGIbXZr/8Za//8KPlBJp8OiUxBZUICt0Eo+RQhkmZVF9IoaEjhqvEnFUmFHTbyQUua0bbFFgjaW2jouLK8kyNJrC2EMh0ErRNg27zQbOIvN2xvXFBcFP6KqUm7MW7wx9tERrw24rYRd7CCWllINzDddXNxTOcrSY89Kjhzx9+oz1Blxpsw9Gw2effsZ2u+Ot19/k/v1zhqC4Wd+Km51zBEnewlhFUdRYFxnSxDRErq+uQE3Ek4jRjrHvUNrge8/UnhK9ZhoGdJXowg7UjBSFM22tokw1LkRedBbZMxlk5yiddAqSZVrXNZOfKAtRApZVcfAr3yeuJyzoImd6irNiNHB8+ijDKY71Nnv+5HxKYa6AseYAQXyxj3wxY2m/Xd9uR0LUeD8xDiNte2d1ehBNIbMVtHjVSGjHvouCED3ej9ypNJV4j0fFpMSC180rKjWjD6NECGaRUsq2DMMwcHlxlYV1NgtPigNNUqiM6lDklYLTs1Mgcf/8DFtYxuBRRnYdNgtfdPY0ku53gATjeMeUkcJusNaJJkIpCfVQmugTeFl4dUyHYbFWWqyBFaQ237P7+Uc+1Of/ByiUtWhjaZqCN49mghZGT2nEEdQVkkm83WxIKTENIxcXFwIj5ULe9z1F3qVorQ85pHKtQlatyvzHOYfRosINMWKVwSotlthZZn86X1CUFRdX16yvV8ybhpPFkhMncOHQ79htex6ennM8X/Cjd99lt9pIMI91pOAZr7a4xtAua9wQaaNht8u5AUpRLxcY61BBmobJj8QQ0UaM8uazGfXbX4IhonzOdLWGcEi6yrdglJYnKaEZSMfdHxbwopBnqbIVlbFsVmtcIWyiq+cXzI6O0S4ydjsZlCcxY6sqCctpXIXCMp8tiOF/Ap27MO8ixspAUFzfJHuxto5aa8qqZehHlFP4MaFDJE4DbZItS7ftqK2lthqrNSZGbLScJEscJowfiFMHccasqtmNkfXmlq7bslwec7Q4YrO+Yj6fE4mCWReWpi4JPjIOPTqLSp5fPsdqxfHxGbvNlnmAq7Dj5OyU0+MT/DhCEgpmnE2cn59ze3tD4QzWRFLomVUlOjV0pgOl2PY9+nmUNJZ5xeQjaQw8XV/mDrIUG11jKOcNWgnLZvCB7WbH5dUl5809bA1FEzCTxSaH73tC3X7ex4UXHnGV8WQjD5c4Eu7Ybr10jtbQNAJbeR+ysCxv/TNDJqEyXTFRVbPDNe1HQCWc0hgV8/cUCb+xhrxv/dx0T73w/32BPDpp+ea3vkIcxjzMfOHn2M8PkuStJlURc3OwP/w0EVPiwYOHBwaHj4H1NBAM7BCBDpVm5ubSZWlRU479wNgPPH/2nN12Bwn85BmHkUpVtMsGrdWB2TJNE/O2ISHdcjNrefXN1yQIzvu8/miclXlJUhqfaYPjONLMZrn4bQ8zCfEGMgzDQAiGqm3xKjCqRHTiPx5DgjGgg9BV/+iv/hExhFMWQ7aDSHeq5J9MUVLijZKpfEpncC3PVyQrYaIfhI0Vx4kYAidHR3cDZ2WY/ETfi311WZaSh5wXvhCmO11CEKdQoX0GhjBRuoraWVAQCbRNxe36KeVouHjyhGkYmen7jG4S/rcxbG8vaKuC5G/5/re/zwfvf8o3vvE1Eh6tNYvlMc8vLrBac/HsKYvqhOiF9rhcHgk0tO4ZwgZQ1FWDGjybzYZ+GNhsb1hvdiyXxywXC5q6QWsIw8DZg/ts+h27YRQ2lVJEDSn5bCPRHe7laRLh1j4ec5u1Ajc3O8ZhZFqt6YKjPp7juy2b6xXdZsdi3uC0YbfeYLVlu14xny2E6vszjp+L4q4Axojxkfl8JiIDPzGGiUIZ+mmLGrbUMWGjks5PgfIT0/MnzBYOrRNHDKhO+tNoFQUWvdtQGUMcYXt1ySaMME28en9ODDuq+Az6W+7PB0JT0KhPcX4fJhxFkKArZu0tRt1yMve8USe03eH8h5THPfW9is5rSnuLbRLTIqG4J1RiPaF0A+y33lfAFSoUnLgB5pHRK55whveRcdUx3u4wToOKnDYt9WKOKzRN2RK7iTRFoklU8xm6tGg7Uc5qnnz2jFM9Y3KBZQ2XF5ecnJxijcKoxMSenZJLqDr0tvkvmZ7WiQ/H06dPmc0agU1iQifysFralH3cWUqickxRfNvlEDhGawVG+nH5zplPnbNc/wD+q+4wg5TEr+X+vYoHpw/EhCzJ0Pcng0yRqGwewmajLBI+iqi2aebsvVv6cWRA9K8GwUP3EGZdlPgYubm94bPPPmMcJ4ZMi2xn1QFLdc6JlXPOnO26LocsiEFVSOnwE9qQKDF4owQqDIFhHPByapmmidubW9qmwWjDer2mruuDFXBKguXvhU3aitpTA17LEC+FvTc9VKWwWCChkhfIIO6laXsY7EXGU6YaxUgIgqXjNBgL2hNzglWKYhFdFCXKFdm62eT35wnZNMsag9H7IavcEwIrZTuDSRaAEAbRkaTI7dVnnCxLYqUIcSJGuPxkZHv7GculotIwXxZ020+5uRqYzSu2m4mydqQo3z/ZxKPXNMn9kL6PlKXl8bMfUlTgmdEPhtJdg96hak90LT5I+DWpxU+KMW4ZxoApNU0BU9hR1YpmtmKatgROSLQkXfDx44/QNkmeACU3qzV1WUtDFMSOvCprXGEJfiRGi7GamBIyP41M6x0pRpaLiiFMpBRY396w221JMVKWFdaI4dnQ93jvca4kTP8T6NxVjKghotVE07Z4rXFJgzKUWqGngMo4tImaECIezbTdME6d+FsrSxg7lNpP4DWVEpOiXT8SNjsRmZRLmEYKBUl1pF2g2wbm9hplZpRxR4qSiG50IASLsTOKdANo0hhprGKIE42rqLUmMWJMwOhImlaktBasMTgSGm1G/CA3ccKS8jMDBdbdwwZPmgRO0D4Q/MSUEgGoRkvZK1TfEXaWGBTHpqHb9rTNDD1GtEo8PD9nDBPbYYefIqt+hTXi4a7oUVgOkMTh/1IiyyLKEE/DvG3587/xp0haUxYF8/lMHPOKEgU5x1GKtTERjcPHgMpClKT2WI8UC43YMJOFM1VpKEsR6ST2fPVwEGR8DiVISiCG/C8q7bFlmdLow+qUQIkcXbjWhsAev1fE7MSosnpTaTEsK9BYEkFlvn4QiqTSjpvbW67XKxkim8hsvjjg3SDFfbPbUMWKbSfukG3bslwu6ccBxgFbWNlyA6O6K/QpRPw0MgYPWjH2njAFnHGMvbgtlmV5GK6FIFbIZVkzjp5x3FFowa7rusIYjSeia4dPOSAlJHQSSIGQMNGio94vrXezCqSrD0I4whpDra0s9El2c/000Y0yPN51wkUvCofRWhhu2y0pSqeqlGQgKMgRcpqisNLp5q79IMzK72WfE3D//us4B0oHCqeISeOaCde0FGqDDiu6ncAtQz8RYw+pYred0DZIRB9g0sR6t6Vwjn6E3Q6KVrHtEsPUMvkbQGBWib0TvFxHCxqG4ZLRjzR5cFtUEaXBujGTK7ZYSgKJfhfRasvNauTjxxW7KVIV4oi53axJaaKu4JWXz7GFQ5U1RlUSGo+ncCNKrWkX0BQDm2cB9IiaWlTOGqiqWthqRYEPUaAch1Auf8bxhy7uOWbvHwOfpJT+klLqBPh/A68DHwD/65TSdf7c/xD4K0AA/g8ppf/6Z7121JriqEIpy6UaKI2hVhaVNLNkaHAkxEAoKenKiXBxeUlVllgUTsuQMgTB6YzWzNsZt9st3ie096QYsUZxvVrz49+LSBkYWbY1f/yX3saWNaevvIG1C559+i7RdzgKjGuBATCYKQoNMcJiNlIW4pQ4TZ6Jgu34FiY/REJhsATjGLYT49RTzSqUUkTjaOoZSTkaq0irb2MLQ3va0A8dQ98R/EhqLB070jRSGkPYKcKwYxx6prpjZW7YdiOJHuOMdPJoiJG2lcT19XrDvKo/R38kd91KaaxM1VAaKqd47aX77M3AUt6OZxtOZIEQZkpVOJRJDMMEOhyWjaQUJLlexqrMcgESOKPuREjo3On/lJv0gL/sR+6CJysE14xf6PwTiRh8bl4P25LDV8u2+W5oO6rEmCYZByfZEU7K8Wy3wiMF1jlHCNVBXl+UBSQOnu6QB2QxMI4DV5eXh6GhmxyFl6GY1vqA32prMLbG5W7TKMekxuwKCU2GMLTWMjj1nqTUweemqWvB/LXGaY0R/IS98Z1JCpPkfWW6F3tZgCuKfD7i5866IGNBYCMlHupTtvPRSmO8RllHuSgYfWA+m2OMZvIjQzb8GydRUcosQmYQphBNgDUWoxTBhzzDWuHHiaPFEqMVZdnw5OkT9NxRNKUUtpi4+Owz4vSEuvA4C1NIOFNJklEUllHd3qOe1ag0YXXJZAPoDqtFNzNN18T9MwloXeLcEQmB+WJIWKfpdwatC5KqBBo2MPUjrqgzFbZE4YlxB2wFKhkDmIhKgXbe0NgjCltRmhkXz77NbO5580sRZz9kd+vQXjHhGLYw+hHaiX6r0LoiDDu6rSUee64uNoRkqKo5IUqm7snpCZtNx9Onzzk9XZL7jJ96/A/p3P+PwPeARf77fwD8rZTSX1VK/Qf57/++Uuod4N8Cvo4EZP9NpdSXf1aOqkqRUkca4xh1wmqx/C1zhzgiOCLKykSfiK5rwhbauiKGTgZmKGLwjP0gnGxj8kRdBD+RRDcNxMZS6bttqW3h2s4wSlMGWSgusGg1ozFLmdAbjYqKoCUsOxL5rJswA1SlxVjFNCmmHNdGEs5vilDZmmQVH3z8CclG6oVFFRF9A4vZMVfbCzozEWKknDYkB6pQFEpk8sMIaRSp/DAGnNFEDde3NywWR5wen/D84lMuri4x2nB0dMRyMUMrxe31islPOG2EWaHU5wQsB5DmxW55Xxhz53vHjc79fsZtx74jHfjlkZQhE0k3grppsCax2/pMxdxrKPd9o3Smaj8M/cKRkEDtFA2kkNWuGdxR6vOFPQkjoa2k5Id0F1E3JilYe9zYaoszlkJpJqXRKRJCxFlDUpbxhdcWap5AIjFKmHki0s4FrhrHUfz3+4ntdkvTNBjbUBbSUZdFJbuClFBJi9AlG2aNwyjivMTn6KXTMNJlyuHeibCdz/GZLrndbvGFY1k1qFH+Xc6s+GUmBbZuGXY97GGzJF5NWmlMafcRKZ+bvaT8vXfbDVfbDcppzs/PsM5QZiFUDImu26KqBoKwtdj2tIs589mcXT+gtKJxBWXhqOqKD97/kLN7D2it5dnzJ7zy8mts4oqvfu0r+G6k61ZoW/Lx++/z9W++TRi3dNPEOO24vLqhbBtM2RIM8vwlw9WoODmaM+xGlFkwjonaOXwoJAO4OSIZC1ExqZEhQkglSWl2vmZWHeHTDVN0zBrhi+t6IEWFn0Z83LHr1/S9pS4NVb3AulPZ7QePsQmtZoT4lDEKm+Xo5CHHD79KU1bcXN4Q379L+4ooxtFTWUfwG8mFxWONYQoKVGAcI2On2axGPvx0y5e/+jWK0rIad5TOset2rLYbfu/3v8dXvvol3v7Ko59WUoE/ZHFXSr0M/KtILur/KX/4Xwd+I//5PwP+NvDv54//v1JKA/C+UupHwK8B/+3P+h6ByC5GbJTCVRhLcYADFYQoz78SFZ1YdYjayyTpbFJ2zfN+QhnF0PfoKN1q0LDrdoRiydGbxyi7t0oV171n24+JKnE5PKUuzlmvH9PoE+au4mihaV3NOA70Y+TqemS36yS1RilO5op7p5UMm7qJGEA8yyGGxGY3UTQ1r731Fk+ffsrVk0tUG2mOCuqU2IURak1hCjTm4IZJEhOpNE6EccRQUdqClEZi0pTljK4bmcYrVqsVKSZOTk84O72HdUUejgpWXBQFaRIXuc/h3CkxjCF3Byp3iJ4pREIQM7WYYQHvPcvFQnj0orbPcIfGKMVq2/ODH7/PFDxvvvEl3pq1WK0OA1iIdP3AMHqsEdhHIJo/gLxDfu1EpjzmxcXs5wW5I31x+GpUosr2lsoEgZAS9DqJ8yMKm5EKpaBMirkSp0ufB8qfbCWAOoWE0YZ+kEDszWYjWPetyh7kjtlsniEeYRhB1gxoTZgmYhQhinECBe26juA9hZK8AW0ddi4h4BqdC7BE+WkljpSFlS7eZNGQzYI1VMI7zWCEQKDUfrgMgcTW7/CTJ/lI4xw6JELYs4i+wFD6wmIfEdqxVRYTILm9WCihvGe33nDv+FTMOkMgThOh22JdTWnFlGsYNfiSwkV0Gnn5wUOO6pqP33uPd975GljLV7/6NX78g3dRrcZ4xfnZGadHD8ElukGsGC7LfM7MAusCwQQmHwnK4to5w7DGEyhMwRAN22HHrt/hFqegFFOcSGULVYNRlu56hZtX7FQE3QKGyQ/E0LPPF1gHR3IVphIleDCG0Vq6sUdp2fHrYCApNklhywXe73BGPPnLqsKYOz+kvq8xOqFtZL0JBOXw0TIMW0ynuV1LZize4D3sdpFnzy959OqGP/3Hfj0THSZC8Nzc3HJ8/oDl8YwHL81/Vkn9Q3fu/zfg/wK8+Gr3U0qf5QfrM6XUef74S8A/euHzHuePfe5QSv27wL8LUBWW2S5Q1ZUMMANM0eOJGA0hBiwRFYWOZ2LCjwNjamk0lGjIwca7ccoDQME1vZUwiDSDqTJs1ytoIkaJB4TeP+kgRkwhQpjQyaLVgqQM/RgoKxgjXN/2bHYj3oPOIQKf7DZsx4Hz8yU3uxXRB/zoiVE6raqsGYyimDecv/Y65eyIq8+esH58y+yNgDO1QEnkaXtW+qWkiCkQk/CTv//9d9ndjvzqL7+DQbPr1rTNjPXtmmdPn1FUlrfeeJ2h77i5usZZ4UivtjuOm5bKmS/IzuXn7gZPRGG1YkqJH/z4PS6vrqSbLSV4wTrH0O34xjtf515tD532SMAq0CrRTxM//NF7TH7iwf3XONgRHhSxSPhySPgkNsBlKarTn1TeE9BPAzfPbwkhUTnDbNYwb1sOWqb9LwU+JHbTeIAolNZ5uCo8/gFRM8esIVgrn3cPkWGa8D7QB4+zBabQB4vdEATiimI3mb3bNX3X07YtIYnAzZVLSFBqwZy7SYKvrbN5MOqkabES1GAV9Js1wzBwsjxGJ0MwUDjDOExoBQaNswU4yYZNKQkbSEFZVPgYCMbiEqCyFbOPfPKjj7i+WHG0OOK1Rw+yzCHzi3OUirrr29ln6VqjKYrETMl9q1H0U8eAh5SIybPza6bUQRYbDd2WtqkZuzVD7PKi75i6DavxM7bdjr/1D/4GZ2cLjCn4h7/zt9j2I+l7HaofOf1kzeblI0geAxTNDO0Vi2bGerbmarXh1Xtv4kp9KHJHxQlV4+iejxQ01K4ipcAmdEQvO8EYIBLxUZFiSUA8Y0JAMhUweYc9yrnJDJ1x7AUWMy3Kyg5zTDpXS0dSe9pAJNiKqm3ZDYq4vuX6+p9yU50zhQ1F4QnBcP/0VzBG896HP6B3UM0rnJoom3NCHTDVLcksaOcn2MryC9/4VX7x1yxvvfVlXn/9S4IaqCRCsu0W7/fq2pr/+3/8n/+EJ0eOf2ZxV0r9JeBZSum3lVK/8c/6fH4ygPoHnt2U0n8K/KcAi3mTVk6xDiMkjdIJh6KO4GxB3TSEYQ1BusjgA2NMDKmgNRIRp4yM3SQtxzCOgaq04vHdOqZjR20dD04e8Hy7A6Rzj5kyp4GsqSeohHE11tSA4Jghyq8f/uhH/Pi9jzHG8eD+Iz744AOGoedX/+jXqWvDbN6yvt2S8BR1i7aGdnFMapY8HT3ry44Y53Bk2T1JfPzuR9gyYWzEWSdKzqxkEx6tmKmpGHFTIHY9Tx8/k5U+34jaWN58/Q2KyrLdbBg68TxJRclms6MsW9L+B727rlJwlcQctu2cOA0M24GjkxPqtqGuGjarNUkpbCX8em0tgmjn9yjNifB674gugumqu1th70eSgHbWUBSW5PfF9W6B/cI9QiIx+Ykf/vA9zk5PuH9+j7Zt/8Dn7G+ylBcdMR3zeUqg2HUDKUF/MlDUUaIQQxCb+LzrM87SGKHjhikeGBgxCPtDxDRCr3PWojNvua5mAtcI9I1GYwPMU81qtWIcPKPvcG3FOA50k+fkaIlG8fy9T/nk8af8iV//4xSFZbVZo50T24tSuP1nsznjMPHjjz+CPIzU2mLOz6lnFdom0YXE7PnjPdOqZ7eZOJ7ru/FpuvPL/+JjqpWWIXQUz5zNtsNoi7GOZTsjaHn2xmHgZLnAFbJ2RwXWJaIPmEJhlCYkGdDu0bagEpNP+CkRpoHLqwuigx9+uKLejbz0T274tj7n5KUlMe14/vwal6AyLU05Z2xLSX0yBmcKtFIczU8wSrN6OnC0POH03inBex4te9arGzbbG45OjrhZr3jerSlsiVUJo+2Bi06+k/fw4/7OFTy/gOhApZw8lic5BzaX/H233YliPSmeXT7FTgXHx0ti6kkpcHR0IkHqxqCMJQaPo+LtL73FrC55erPjKy/PuP/gJU5PTpjPlpTlDG2ceLzfXLPZbNhtu2wgWLJcLvFTkkXsZxx/mM79TwD/c6XUbwIVsFBK/T+Bp0qph7lrfwg8y5//GHjlha9/Gfj0Z32DFCN6N1IrBTHIAIqEA9p6ztH5a3x2cUEcPTEK9vmDxxc8Xn3Ka8uKr792P7M4FLaoCVi8CnQoTNuSjhqCiex6TxxvqYsSUyhUNtRXSWG0Frm5NiK6SS1WGWIYsK4lEYlxomkMv/itt7l3do+joyVvv32Pm9tb6rpk6EWA4oNHaVBFS310hi4arsaJ7/74U2I/YhvLSdNizl/mpfoVrNoy+A2ukODepBLGKJw1DP2AM46yclCc8OoYeeXlhwQiaM1/9V/+bbabnn/pz/0xlqdLrDZipWudLBaj5+piS1k2pPRFXqz0bzGIcCxlTx7rHInE0fGS3W6LDwGtDP045ZCEu1dQgM5MSuLdgDNGL0PUvFDt3T73/EulOJAm5cPqhVc8jGYh7dWqCeMkRHnPId935Yd6le48tNX+eyH0PoIQt+N+oKkUFRqtFKtxZOx6UXpOEVtVFKWjsBUxTqzGkZQ8xhVYLa/hrAQt13WFcRqTEiYh6kYvw0qdEqw71qsN227k+mbFS6/chxA5to0EU+w8jNJ06wS+n9BDYBzX1G2DcyIeCmHAbweatgKjUNqgQ6CICuvl505RKH06eUpXMasNGpuReKEo3mWu3i3QcupStt6VnePV1SW3Nze8/aUvcXb0CG0tT588wY+aP/1rf46iKLi6uGR1u2Hntrz04FWOzs5YbdeUdUlUkWHouV1d8uppQYiKsV9zsXvC8eKMaBVTmFDdNd1Jw9E6cFV5blfPuPE92sNxc8xqXBGt5tPn7+MKh9aWumyoqpqqqOjDBNoSk8zXZs2SsYfXXv0KwY+Y9Ixnbs0rZ++gFDw6eYtIIKiRceoZfMc49cQofjcxBerS0aolDs1Ivt9VQmNEaL1nnaUgViNTxDjFfFmxuukoZhMn9xq8PULZyGeXt5RFyUsvv0zVVjw8e8BxqSlKz5fe+hW0dlL4k2a3CVw8f0rfycC+KB3z2YL79x9S1zVF9tXZp9T9rOOfWdxTSv8h8B/mB/A3gP9zSul/q5T6j4F/G/ir+fe/nr/kbwD/hVLqP0EGqm8D/93P+h46Joptjy73XWHCJ7HDjVqjyoKkDaqsiF7jA8zuPeLLj5Yc1YZNZRFPcoUqJR5PEwlKmDiMEHYju25k5y/51W+9RVu3eB/psx2un8QaVnBKZKWvLEkJB3kvKHnzzdez8lN8bObzhrat6fteMjV3HQoxJ7saInGj0aonKTB9ZB0LXp2fM/mJ731yw+7ejF+9v6AtWhHstKdoo/FB6ImqVqATlS04/dLLLG0pQi+dWG1uWV2tiMC3f/s9Tk+WfOXtN3n1lXPadoZzlnGcWF198MIF/dxv+RrfmTmRYLvZUpSO7W4n6TFFwWxZMOYcy/307ZD0kznSn39N6agPBgG5ORdWyXiwlvgJdxwvNvJf7Or3Ss4Xv5dSmhQ1Kca7AS17lowsEtM4cbta8+orr0kYhjYc2YopTdzc7EhdTyRwe3XLop0xn885Pp+xGgb+yT/6x9y7fyLD0sKxPDuVrXISg65yDEI9BFlw8jqWdGbax0iYPOubDdvFjLoqUUmzZ40qpQ9FNyXohwGbB6w65DmDSmgjAc1aqZytSi7qCdKL7BeFMZpZO8coWexftGRQL/z/84fsRE5PTzk5vQdZhZ2iKIXHYWSz3hB9xNWO7XrLxdMLSInTk3vsusD3v/MBf+pP/UkUsJk2/ON/+nv8hb/4F5kfLfj0ox9x++kNj07e5Pj8PlMYuV0+44fpxzx45RHL1kGKHMWRj3/0HuWs4rWXFmymEeM0QW2JcaKfbljvnhNSpFx4unDBdH2LxtD7kqJquf/gER9//AGkxDR4nC749LNPKUvH2fkZMQVJgjL7fkNsBEKMhHsegmIzXtD7nVCTY0+IEz5NmWvk5bnxAyEM1G2irhzNAgwDTbvkq199h5N7J+w2FUfLI15+1FI6h9WO3eaWru9QeqAfNozDANphCxF9nZ0tqKrqc5YT+zzdvWDuxWv6k47/MTz3vwr8NaXUXwE+Av4yQErpO0qpvwZ8F4ng+fd+FlMGoFnO+fq//CdAaZQyAn8pckaqY2LH/M1XGIaR5588ZdNvSIUlhIGrrWa86YneE5Pw0xMw+QAp4qdBholorm9uuXd0zBuvvUQXK8YQ6Hth07RtS+E0ftOjrQRil3XBZrNGrLDFpW+aJsI0Er1CG4g+ME4+D6syK0IpFAYTRqbNU54//YSj03Pa8yM0ilV3S7daUU0r+qsLuvuvU2kZqBUJSBLQvGdn1KVjWVUUStE6i7VygpaF4X/3v/pf8N7HT/jBjz/k5vaW3/rt3+F3f9/xxutv8PWvfJn5rOHs9BinNevVwGbbU7e1LH9JZZ8bz8XFJW1TolCs17d0lwMxJZ4/e45zjsvrS9brDQ/vPyAog0kwTIEpGkqjsIjkXau90wzcrSQ6dzsQJs/F0yvaWc181lAsm0zEyf1k3vKmvXl3FIM2nWRnBdLhim/5HcQQ47447pcU9UL9Ugx9YHW7EaFNTCiyoyaSquSHiHOWtp1RZqrhPjgc5bCqRAVoXIHqeqZJOM/1XIlNL7wAS+13G7JpIILVisIU+CGiC4WKSVhfCUxS6BjRKUJm0xR1m1Wd2dZXQVm12EJmSJLZKcPY/fBd1oAkn58iWvx1iTHIn1OOK0TaH2EjQSKnY6mQd0pC3TQoSJ7zkyW2KJk2KyoFp/OGqi6YVSXX1pKSwRSW49mS48URTd2Cgrpp+bVf/6Oc3T+nbmoun4mFb13XnN27x67boSO8/+HHqHYhYrCkCbueH/3wGe037/Pq+SNaJddB721095L/DDMZo/HB48dsaZ0MP3rvKT5M1POaL335DWL03D8/IyYRtKFkkZRMgIhW4DJd09gGFSLzpiGkQDf2DMOQG7gBpcE5Azpi3zzHlYa61cyaGbNmgRprjo+W1EXDFALvbj6lcpoUFbvdxDh1mSxQ47ueqqxZzuc4V+GsO/gI7Yv4nYurHHvNxj/X4p5S+tsIK4aU0iXwZ3/K5/1HCLPmD3X4ANc7y2p1TVWWzJenWJel11Gscau2om41v/+dj/nd3/0ORWFo2pa6XnJ7e8Gv/PIfEdoj0M7mXF/f8tknj9msbvmX/qXf4OVX3+Rv/e2/w/XFc5anJxwf32PTdVTDkHMqU/budgdL2LIss9NdZL/hT0lis7Q2hOwL3g8jRSmUOa31QWp9+eljnKuYDSs++p3v8pV3vsFZIc6SqUn88JOPWXdb3NdexzhZoQsn8udx9DjnKFxJrSMzqymUPgh5UvK0leXXfvEd5m3Lj9/7EReXV5RVTWLGuz98j3e/932+8tZruFnD2bHc2LerNcf3TsQHJN8w87bOPh4BhaT1BB+w1nL0xptU1Z7nrTleHsvgmSTD0ZiwyoAl28vCgVfHH+wPj5YzmrrMtD6xmRAc8y4oet+s38EtmSqo1GGA+sXDWpvTa/Zf8nkV63q9FVe9mA4LiNgPi2WsMUJ3rU1FWYr1a64fWCsGUQCb2xVN5QjTiHElanaH96uUIcYMBx3ea+bmS/pSz2Je3/1giReopFHEQcZIAHMhxnVHe9/uJF5G1rqsDLX5HMcDk3S/f7p/dkykxFqwNi8yeQXYv7M79QAvnJN8vpGCYq1ABmjLa2+8hbNynoqy4Ovf+Aan9x7w3/7W/5dxHHjr7dd4cH6foirFYI7E/QdnIlBTCVc4iqpmsTziaLlgNp9xenzMo0cPZWhvLTEpYlK8+eZb1HXJ48cfc726QWl18HFJWeGwHwq7lLC6YNt3VFXNMMB6t8bYyL2TI156ec7Q90zZeM450SxIJGdCG9mhKHR+vhPGJMZpkLlNipTKUFUtoZRmxFmH1vDgjUcohB6rjMaZEjOzOFszRJimyMnxEkistwFjLEVRUxQWY0vRP2h96MrlWuxZTXcspf2vFwv7v8jO/Z/bcbQ85i/9K/86v/Xf/13my5q33vylu6c733TTNOF95O//nX+AQjIZu93IZvuUMA38/u99l7K4L/ii7um6gaubJ7z+6JTzo5ZSez58/z2ur654/uRj5k2JmkaUH7LAxwIWXydcWWSWnWS7ej8RoyeGhHMFzpYopYlpYgqJUluKsqKq5GJUleBiElbQ8OmTJ2ASb7z2iuRCWktMkYePTtl1nTBSdJIAjpTwmWXT9z3GaAZtmaZE4xSN1ojj+D6jVPPOV97k9Pzf4ts/+AHf/r3v8uSZ8N1ff/SAl48WPL265rd+9BEPX3mFf+3f+N/w5JNPaCu4unxGDJ6yMJRljQ+BzabjK19680DJa5rmc10DACmzWyIZZ5fJmVYwm80JPuSknDu+eG4scUZRz/ICJ2hDZoFw+PVi+dbGMFvOee3NVynKCm3d57D6w1t6EXZ44SGRP+SgaaWZJg8pELLYiT2en6XxKI019jBwU0qiHiXjVJH8nU7AHPJIVaa+3n1fpfTBwnVvCVvXDVUpwzc5jXcWDuRiOk2iUG2ahsJYtqs1nD/IkYTCphKDrv3OJoLhc9CLUopZ23xhpmEpS4vWRpwIs/nvvkSmBCpp5rMZxpWgNffPH3ByeootS9BWdk7551VKU1ZQbzaMk8AZ2hrqtpYuOkyEFIhekp1QsgC9/sbbLI+OiSk7JVpHa4tDsVLKEJOcq5QCJ0dLGiMd9ocfvM/y9ATjamJKoHOoibKEccAow/bmmoubifnRnA/ee5+mfoPSyerWDRqjxKGvKOQZ3lNZ+2ki+MCeGg2KZC1Ka+qiQisZohMR584Ys+lbka9FRGtHUqISXvc5u8BYFsdH6Bw47nJHjoKUAjHlZ0CZfEllNkIgp1jZwzVVSjRpOili/EktzuePn4vi3vUbfv8H/5DHn77HebjHV7404FxBiJGUsjxdR5QKTNOOd975Cp98+inONQzTjugN89kZ94/+NDGJv8elesbzy+/xS7/4Lb761ut8dnEjzBKt6G6umbkJZyYK61GsmaJijInNWglft6poZw3G6GzaNIJWmY0RxVdkkoCBwjiMseKiqEVunVLEOglpfvDgAffunUtSzSRiJbSoHV0h+YvBi3+HRtGPI9Y4yrLk5uaWqqqYzxfsPAwJGmcodbzDnlNgXhf8+q98kz/+q7/MJx8/5bd/57exKvEP/5u/z/2XX6WdVSyOFvRTz/Pnz/gnH/yIe2dH1GVBO2t4sdqKb1Ykpbyd564jPCC2SomJV0g4J+Zi913Dn//zfxZrDFVZUFUFMYtw5Cv37WX2QlEZRjms4Z/vRFKShJ+2bTlezPINfQf63A1N5f3oF7avd8VC/m5zFGGR8ze7rsN6j7aKFGVrv3eqdM7lAax8n7aR+YW1sqC6QhOTl4T7pO6w9ryYqYzHKCX/ZrRGFyX37rVUlaVtRKW8/2mnccw7BOmKvR+pG0dTl3ztq29LklEwPHpwDCphdM5NUqCy+drnZxD5nBwumcYVlvv3H9LmQHcJoJ8Yx4EQEmMrwSF1XVGUFa4sOTt7SDNfgNKEF75B3M82UmI2n/GtX/xljo6PuV1d4L2wkFBakpPQ2THT8OZbb6OFWyissIxlqWwwITu2RNyPWYTtgHECTT1YtDRNgS4LhmGgKgp++O67nJ+d8/j9j1ltO1596QH/9IPv8kf/2K9z/eyC+R/5GrO6RTWOT548ZtdvmM3u0fdbJj8RkQ4+KtnJiuRfCrmzFpVnKeMwcn29oe97EZiRpEa07cHO12pNoSXxba8u3u8uVIyHnanZZxbHKOyiGFEhYK0w82KKAkFlWvT+90QiTR4K8dL6A03XF46fi+KOSkTnef3LrxNj4Le+/Q95+uGnWAd66UiTJyQYr3s+/fQjdGkY/UjZntE0pzgr29ir+F2SksHPJ89/RIxjPjEjT7eP+davv0myivWi5x988jsEtQ8QBqUNNS3l+ADrBbPXJlGWJVoLa2WzXovPtAFrZeDY7TrpLtsZY9/lwYdspXT2kY9B/MQlLk1n7HNPxcrmTmj8JJFuxmhG3zNMHW09I6TA5fVzFrMldVlxO3kMisop0AobwSiFxVAWBV9641VeeXTG84sLvjN/wNX1NfePS+4dLdh2E2XluLm+pKgs3/3eYx7ef8jpvXOxc9UWg+Fe2/Ltf/x7LL9UUx8do7QIcZx12bJXgjuGQbzsC2tplCGS0EbnDlO64JPT87uv1Yq+30me5dgfFJhk9Szqjh2jlGzDTYYdTKa7xrTflt/BPzF/zt3X7rtj2GwHtLU07QKjpXOaQmLqBpq2yLx1lbfHUoz66Y5ZpFMiTjn8WGW8lSIXoJiDwV8oqC/goacnxxwfHcs8SRsZJGdbYlLkaLlgHAecNSgkBSjmzt8YjTI5n1Ul3OFplTHxvnjfJVDJL600xycnOFfiCmkSiqJkNpvd+YonxPUx3nXud1ia/CzGykITUyDExDglofgFhU8jzko49ZtvfRW0JxFp2wVaFzkoWl5fJwkD+WIxEm3CvlNGbKZz0PUURoxRrFdrbq6uWLQN//jZY9pNy0v1Ef/g7/8jfvMv/Sbf/v6P+JqasQ2By9WW4xPPME6AmGydHB9TOUsMkQdnx6QYOFs2NFWSvNgowRlBFVhbkJJmGEd2w04ycodeGFI+sN1sCEEsDRISHjONk4SwRPB44tAdLJxTmGShUEW+R1Xu0uUaGLOntepDHoAxhugjyd3BLvsCb7JvliA28ZBM9dOOn4vivt6t+fu/+7cPBU+j8c936EWFmSxBRzSJpbcoPFMcKOqKb3zrV7H1kcjDs5WmNoaby2s+/eR9FDDFwLvPPuLp8JzjV+aAOBVOwQtyJ7s7bFRYNRPr13y7jePAbDYjBLHrFJaHdBjTNDKNYoRUZQ+SaRpFNZsLlolCsSyrgmEQXH9PEdTWZmh6rxgUnB+19zIUDHu1WeGso61bhmHA+yA7BFeyHiamKTC3itIUGI0oN2PEKEmyv7GKN77+dU6KxPc/fM7vfO8ZX3/rhLKy9P3I0ckZT55fsjx9wD/6B/8dv/Fn/yT9bqBwlo/efY9f/OY3ePXtrxwojXs/971PzR0OmAuv2t+IYkJ12PK/QHWcpyNISVgI0yQB3n3PerOiKKZD55SSzje+OOEp4gG7eZH7keAwzN6/t33BS8Cm6wgpgTb0Xcft7Y0MVXOnf7ScSQeXt8taKTbbHcfLFkiEydOHRFuXaAsheCnwSh6fuB/gcqiLh52DtXtjAHVYxA7BGsDLLz/ilVdePnTtWkuRN9pRlKV4docIjPm1VdY9gtKZwmstzhUUrqCqKsqq4uze2ece/v25OQz8RdiRNRU6Dyc5ROGhFOM4cHP5jBBHisIwTg5tPHFKDHGLM3MqF/CxADOgVC3BMhQC3WUDviefPeazxx/zzT/yK7noJ0IKEs4xyvXu+54PP/iAhw8ecHt1yd/9e3+XP/Mbf4rHjz/m2//02/zP/ty/zI8/e85yNjC7V3Fz27HbeW5uNoCl70YwhjEE0WUYJYthikQCvZ+IccQoT1u0VFqoxGAxTUNEE0JiO3l8Ek8ehdzD0XsK51Ct2I5vxi0ohdGi3RCKKRDEPE2zt7vwMqtJGQLESspUSAfCiFJ32PoeZ4c8Q8p2Ey9i7TKn4vA1P+v4uSjuKIVKgl8mDf2zLf7ao17/Kq/ELavpGafNBaWeMG5A2RKlHMYJVejq8pLbm2ecnN6nXRyLMhFIKvL44hOGoyVTLqIWoV4S5dHXyHbeKkuKlWyh81RpmiaU0ozjjq7fkZDtkvd3XiPWCsMihHjwIKmqKv9YOdVos5HXNFI8QoiSKfFCclCKIpG3Th+61z1FzoeJ1XbF6VIoeBeXF7RNzWK+YIqJq3GiNIa50xRJuv+iKpi3M47Kgspp3v7SK+xCwcOkqYpADAMhzHj27BnGSEhx0864ulrz/vs/5uG//Cf5Y7/5m+jZEdttzyxHBN5dsrtQChk25sFPzDBB3lKTqYsvHnLDJ7QzOFeJFS+Js3j/UHy01pJakxIxTozTxND19H3HMA34SO50LUqJcdeePXGAJZQUwP1gFCJX19dUbcM4jJR5m3x2tszGZgofJmIKnJ8eoZDdwmuv3MdYg7OiFNxsbzDaUJUNpSsOhXovF9pnZ7KHmxCR0R431VriD60pODo65uzsHOsKjFH83b/739ANPQ9eepU333iT4EeGYWS93ua4wSSJSK44WALvE5GUUhhtmaaRceyzPe9dQd9rDQ7n9wVUfr8Y7WP8QoqovNO0upDcWwNKW5LyqBBJMeB1zDqIQPAdpSsRK6CI1rI4hTBROMU0dWxue66vb7h3dsrN7TV/82/+Tf7Mn/4zKBR/46//df7Cv/KvUJWWTz/5hM1K0p36Xn6W8nKDv+nYVTNiCpSVIxIorJF0tRR5eP+Mr49v8dL5Of/mv/6vMq8KjAoYmzgqJ4aho9BHJMRVNAFTCkw+MU0xEws0R8sjjDF0WwmpLpwjhZijFX2GjsS62HuNNS6fSbkXldGHc7/vzmP07ANMVA5bf/H8H+jIis8V+v2/v8ia2YeO/6zj56O455UpJUVlCl5+8AqP28h3YuR+UsyJ6P4pnkCyMrhDJZSR4uIKR9Msca4kRfFgV0ZjjGU7bDiJlTxcWgv3/bAFVajscV3rFqacPo/KwRRSiK6vb4QCpaTj6PvhcJEkbkwzTdlHfo+1atnSyg0g4dqoxDj4PAiSrv0wCc9sBmlMlaQHBaFpGW1QBuHFVoqmqRnHkdvbG2azGbZwDD4wdCONVcwKhwOqoqKsHGHqUSngjGFRWlbbkT/267/CeuuJY8flzQo/TVxfXzCNL7M8WfB82/H9i+f8999/l4cnx/zpP/VrbDdbXn/9tcN73ncOKVs+SphzXvS0Rhd3MMWL50ZYL3dOhnu8HPP5G14pgZuMLnG2FMvm/P1SxiYV8PzZU1Y3VwevnxcZCFGwGlJIJC0hHZvthvV2TdEKo4MkfHWlFdZYEhpKlQfDmuPjZs+VyowXg0JiEvcDYeAQ+EwexFlrRFlpnWRutjPpBouasmlzDF0ecCpFil4GxilQ1C3t4hiTUZQHyBBV8Gz5mHmhc5tyjq7WNtsvj9lil3xt4h1WfqDXpczGiIfCP2T2WCQx5QD4uq7wU8xAmM7h6S2Fq7FFwuT0IpW92se+49mzC46Pzxi6NRfPn5Fi4t0f/oCb2zV/7+/9Pf6dv/LvYIzik8cfs7q9pq1bdusVUz9gmOi2vVyLKL5Gnzx5yu9/9/tUVcmbb77OYt5iNPzar/wRXn5wyjfefoBH0xY1b73+KioFjmclKU6kJAywN165L7bOITBGx+g93dAzhYiPOjPlxKY4xpT919OBQVO4Au8lGSspgemMLbGF1BqtJBN4PxzV3N2Las+gyuZ3wXvUFxgyomHQRBU/F9sId8V8zyr6F81z/+d+WKM5Klr+/o8+oI+G3zh9m+24Y6aOUOkRU/qEFBPz+TmbXS/4qdE0TUNb1bLNMjZbq0aCv6Ng7B9ElEAXKndVSikcBhsboSOxp8AFUhTORNd1InwxKgcXp88VkH2MVl3Xh4s1TaN8X6AwBYWzzGYtUzWy3Q0SbJEhHoV0hCnFXITU5wpmjB4VFIUKDONIP+xYtHNmTctms0UpsTc2ztEHz9iNtEWB1QUaxawysktJA85oZpXitZfewGjNO2++yW//0+9QNjNUnLBa8+TxpxyfnhGmwDhOtPMF/+R3v82PfvQef/nf/Df4/ve/z7e+9U2apuHABcfgveS/ppRI2URJG6Hb7YvLQW1qLEYbrMmsjcRBQbk/hId8t2FQOlvmaqEv6qRRWrNcnGCUoR8G6jqitcIVNtMjE88vbqR4O0NKgWEY2G23nC+OUEkfOCP5u2SIaS+BusNJjZY8UW00IXiMc5w/eBlXCMZaFiXOFbSLhWDJSsnia4RGyn6AjEYbB0i3hpL7JCaZ5VhXi+VFCqSQAMmxVVqLEnu/M8lYrTVGdKgZHgve03c9iiTv08jn7TtIuX/vtvf7MxyCFBUfggwFEapo4SqiSQyTaB9c4Zi8EAMKCgyRv/t3fovTe2d87Z2vcHVxxX/+n/0X/C//jX+N5XzBbrslacXutmPoOrbrFZfPn2Odod/tUEBZCGNmMZ8hSgmP0omqKpk1JcNuwy9+/etcXl7xi1//Br/yrV+kKgr+xK/8ihicackf1cmI+DEFYhwZRy/zlSkyTIHd6OmngEdsJrQqMVY68MkHVIg5llEYVdFPct1DPAz8tdaCA6RIDAmlLNa6vPvbW1MDWjIL9qwcpxXionoHf5lcqPfXB6CwYma3zw7Y1xmBaHLM5QvN1U+tp3+Imvv/lyOphDOazfWW9z++REXFkTnCvjlnOTshxDPe+2jG0H9Ee39BNwwYEzGFIYaJGCU4eZwmiXRTe7tY6bZMpu3J98odo5EVtdRz1FRgrHTX+/xWrcgUTI9CS/BtlJPeNI281gtdbFEUWGtzqvpEmCRgd89dlgAQB2mUUAijiXmb7nJg7546orWs4CDRbdoYKGVBismz3W7wo+fk+JjNZseTp08PYRFKO27HCUVkuVyymEn79+arZ3Jj+QodE1YnXGH547/yi3SDZ978RW42a2ZlgY2Kk/mS7mbN1dUNZe04ufeIJ8+v+Dt/7x/y6muv8Tv/5HdpKssv/MI3STEQYmAYRrnprdDuTNJM3ucQjbuuse97yqogJpdZBHsmEp/rWowxubhKcMeLhSklRfSRum6o6+awgEyTsEAk6SfwysvZGlVrPvr0ibg+xsgbr7+FTiKT71/wZ9+HYIBiNnlMvq6Fc4zDwMXVU2IILI9PeOWNL2EKJ7sVI911WeaMV5UyxKfvaJcxyY5DycM8eS+uhHFk6HrGEHnw6DW6YcsPf/x92lnNYnaC0QVVWUt3rvLOlL1fT8yLY0JnKHDsB3yYZIHVe5ql4Ns2UylD8HkOEPPwMzs/KuSahURd1SgUH3z4IWVV0s5bfAz8V//1f01dtfzmX/wLgOLxx5+yWfV89ctvMw0Dfddxc3VFXTrKsmLnR2aLOYWSAfW8bQhhgig7SkWiKgqshvlsxp/4o7/Moi6497Wv8M0vvcl7H3zI1956iw+spS0LylKTfMBgMFZhbSVGb0Hort0QWG86truOwQeCktmEsbXEXwafee89RVEQlWKYRK2eUsLtzdJjOMRrxgxf7nY72X3nBs/HwOgm2TXloahAw6Ji9zFRWIOzhhjA+0xxzCyYPba+v/dQAsWGHA3mJy/zFWMwWmpJWZafCx//ScfPRXFPCpLVMrwwiT/5pa/AGIjPtvzKH/9l9PEJv/X9Nde7kiHsEF6IRifP1dPPePcH38fGxPnDB/zohz9CG/BjR8y0OaUzuq4gqj3LQOXOWmPSjBgMyu0pZiK5lq1WftBTNqJSiqquJdBgkk5V7R8YJav5/kLEEEXBOooZ2Hab4YmUB1x5J7HH2vZbN1JCKyMdYYhoozIPWyblPkRSnPCDx0+esqxYLhfc3Nyy2aw5Pj6hnbXEGDk5P0dpWEVH7UpKI6/vwyQDHRJJaZq64MtvvkI37HjrtVf48LML+mkEJrpuzWxxxgcffswb/z/q/jTWsiy77wN/e5/xTm9+MUdkZEZm1sSaSJrF4kzRoizJkmyIsga0WrIEC2gYdvc3yf2xAQH6ZMAfW2ig7Ua32pLdVku2ZMpti5JISmKJLBXJKlZW5RQZ85uHO5xx790f1t7nnPsiMlkN2I30ATJfvPfuu/ecPay91n/913/dv8NsYwOnFN995x32d7e5ffsev/Zr/4KvfvXLbG1u4CxYvY7hSmMQed6qrPhn//Qf8+Uf/hJ3btzGKQlHjU+KBg9TAdaIDK7WksCK4rjjzIeQORgs5blmWRSR5SNmG77LkD8sWms5PDljWdRMxjPu3H3dJ7ERjX4PATRN671+Mb5oXzwURTRVKUleaxmPJ2xubEtCVYtGD0rmyylom0qUCNtAO2xpmopVscQpYWKtfBcnod5lfO6LX+H6tZsUxcIfCDVt67DaoHXjk/YRTkU45/WH6kaMgzMkSijERSWeu9KKPEt9oV7IP/jcCPJva8SJsH7tnhyf83vf/R5f/fJXSNMRVV3xy//oH3Hn7h1+6qd/kta0FGXBclFxcT7HWsvZ2RlJknF6dkpdN55FkpGkmqo1zBcxWxtjYr0icg5bV2htefO1O+QRbIwz/tKf/7PsbG6SxBG/8DM/RdtIj1GU5s61HX735H3yNKEoSpJ44udOUxhLU5YUZcFiXlI2wh+P0og4G5FkDtU2ct/FCmNsz0bSilWxom09V9+ziYLMRWslijGtoW4bScxHAscqJXr+YpxFurqjLzrnKY0exrGOsqgwBqF5+8TvOrbucC6iaQzWGS9O55uxdMyw/jOu4vJXr0+FcVdIpGnamqdH53z07SPKsmZ15xY/trPPbO8W9p1vc//ONu/8jpdTXRS8+6/+Oe3zC3JnibKcwxNFVGwxuX7EMm4xmfCWxYgHzlwwDRLmJipHmRStuvwmTd1Q1SVJmggP2OPpbduS5RlKK6q6oihWANLRvBUJ4cqKQRf4Rku1noLWWpqVvD6Ko96r0302XDatgdYi7eDwr489DKRwVlTtjFcAdM6yWs0pCtjc3KaqKk6Pj1gtF2zt7pLmI9rKcLZqWSWaWaJItZSmh55GztkO/0+ihM1xxq3dLYxpuNwYs7u7T1HXWCMb9vT0lKIoUTpi/9oNLhdL3vn+e7z92bdpbcs3f+u3+emf/HHGns8N3jNHBL+U1ty/+xrjfErTWoEPlOQ+QiRkjMEZSxRZYTb55HVsjcdiewinK/TQw4pLMVSyiXxi1QdG4/EY5STpHKc5xjpSpTpMM3VhbJ1n4vRYtUaho0g43ErTtiWVrTAOilVB065ozAprHXVp0JFiZ+sm5xfP0WpEkmqyPCdOUra2dlkuLyQa1DFRlIBSJEmKVSM/z0FHBBpjWJUlSRJhjIT8o1HK5fkFSSLtHpNoJJ24dIRta7SOqNuW2D9PWZbCO08lYmqahn/wD36Z+/ff4Atf+jwWx6os+a1v/Q5vv/15phuOi/mcxWrFxeWci8tLlHLML+eM8illuZSo1BriTOodilXR7eskSmlag3UT4lizuTnlD/7CTzFKIc9G/Ok/+SdEniGC7f0t3zRcMG7j38SiieOULM7Y3tjDupiLwlKZhqq5oG4N1kpf1yTN2ZilJIn00l0WSy4uLtBacnF1U1OU0mJvMpmQRAlxEsta8ZFenueMRyNqT/N1IE1zGum6pr2XnaYpKEXT9rUc4XK+yXiU+ObzbUOSZMSxYjbboGlayrru1lVQTTXGomjAKVQccXl2wpOHH3D7zl2mWzuYKGgvqf+VwDIOVGOpsZxfFBydXghTYWuDyfYeERpblVzb3SP2O3RVrHjy+DG30h1mswmlmjI/nrGqoDp/hnWGONEeJuuNhlLKNwSSzvOxmmIqpPAIwdqbtiSKFJubm1grHpa1lvF4LLQtn3TKshyQBEdZlp0hi2IfLjsjgoTGEMWi1+EQTHgym9CalqoxaK0QRMYbFRTWSIFDolPB3Y2lVS1x5EuWfbs7lEFpR93UnJweMJttcPPmDU7OTjg6esrGxhazyTaxi2mbmtO6JU9T8kgxiqTBhVScBl64Q4qict68eY0HN64xmmzw7gcfkeqYTDkwLUmksMbw9NkTdnd2GI+mTKdTjk+PePf9D/jxr32N+XzBN77xTb7wxc+zu7uNzyKTRDHbn3ubf6kaftI5ZkIlobWuM6Qyax3zugtfneAP4h15JkxH8fYNRMI829Z22KhOUpbLJSBUxtWyREd+TjxrJkKE64ZiZRK+O+qqoGlaLs6OSfMR2WTMfDnnu+/+Nk1Uc/36bVZlQWtKRumESMHW1iZxotnbuQ6qJVIj4gRQEVonxDonzxspdvFMGKWEMtvWtUR+rkIr6b2LiijLAlxG0wBOk2cJZV1hjCKKLco1jLKMuq5YzC8ZjUfMZlNpwm0s/+0/+Idcu3aNn//5n8chTa/PzxccH1/QthanLBeXlxRFzeXlnOl0zHgypmlqsjRllOU4JfOQJBFppqlKX7jkE75xHPHmW6+TpwkpcHN7i8Xlgu3ZFtcmCXdu7vjo1UozHedwxqGcSAxbhVdutL5pTIJNRtx96y3qBk6KAleUxGlOmk7ZmsTEcer1diISTyk2xtK2MQpHVdUYLE5DmmUi+qZjssQ7a1XVMd2kIbXkLsKltLSubFvZy0pLEV+UxNilOB460r5uQNajOK1Spepcn3OazxeCs3uHxVpLVZYCL2Yp3/7db6BsxBe+8MOcHb3gd7/1a2Sjn2Vr/2bn+IRo4ZOuT4dxB3CwvbHFj3/1TR5/5+8RO/jM7VtEcU5rHU29Io0k23x5eUlrDLVtKfZSpte2cU3DRn6MWsyZz+fS1zCSEQ4629Zzv/Fc98QmxG6KVQqnoyDnTpyKzGqaxhQrgWKSRHDVsiyp26bje4N4PyAGKLxOK42OEK8g6r0k53m3bdNy49YOT58cYtSA2hQUBROLVhFGO19N2JKqCBW1aOWbDOAwziGyXQ0Ox8XlBcvViu2dHbSC5bIgSzeEC53nwvaoaorGUSWKcQqJgsgFPx6Ms55eJr1ZLJbPvn6Tz92/zsViwU/88BfYHMciPeAMo1HCxVzG3baGre1tstGEk/MLfuMbv82de/fY3t3m/fc/JEJx58EdLtuCf15XfGU0Y0MrKcW364waiUR7g98lZOnhBKU1tm3QkYiHaR0xTE6FfEjRyLw5Z6krgRVaa4hs4/OJcjo4HMZXfTZNi8PQto6yPidJcopmwdtf+AqJLzbKRymtLbm2c595ckFrVqTRCKUNOs4Q7TzIsgzTKqxtMMahtSVLc+mVa1qcMyhGOGpGuaEullR1g9YNSTyhaQ2xTqmr2pfBi/AV1nJydECWxGxtb+DUkqZcUlQF/9Xf/bv8zM/8NF/8oS9gLLQOVnXDyfklRd0QK8d8ueRicc5OeZPlqiTNhKIJiulsynTqBcysJY4gjpTAQAo0BmcaYu34+o9+le2tLVLXkk0S/sy/82+SIBTZdJbwnWLO2XnC/ngkRTpIylprMeYORWMUrYKybSnKivmiom4MVdOgIk2cZSSjmFT1x75AcqnPc0jxl6wXWTOR7xLWNI20KInEGGiniVTEKJc8Ru2rhJWKyPMRGkddVSLrgMJpsNqL/SqFcZayKdna2mQyHdG2huWqxDiDiNIpX6kakXndqdVq5SPQQB/G5/OEaaUwNMWCFx++i3IRn3v789y8dZ0/8LM/xezmaxBFRH49R1FCaB7zcdenwrgrFG/d/ww/9rmvszx3/I9//59i24bJdIpHhdmeTZiNc9IkYmNjg+RIGv0ulxcsH12Qb2iiMexuJdzIdsjSfb7/vceoLMbF2tPmVH+iokjVBFNKWN/RvGJhJhhjSZKU0+JSGARtS+WTLTp44S4UfAibQodCHiNeZZZmnYESmpqwPSItiZlnT59LSX1QO1Qa55Ooos0SNLoDc0RwPBfEr1wkk2yAQQl3XVsOD06ZbUzY3NwlisXjK4qCzc1N8lFG27bM24plAbM0YhJpYif8e7kD3ZXXg0Un0pziWpbyx37x56is4vMP3mAyyYlQPkoRxtDJyQmXlxdUVcF0Y4PxZEJjDL/667/OLB9x7c1bzBrDzaLAGWinm0gHbtWxOa4KJw1lBDocDzycpWnaBq0iiqLoDspQACLjpqQWwTjOz89YLEsu55foSNQgjRG1QOMMTdMwmYyJE9HEjyIHTJmMN1nMlySpqPc5p0mSmNZVKFKSOMW6Sio66wpah3MN0xEUxQpFBqEqGgnVq6rCtA1BL9y4sgvPrbG+YlkKpl48f86quCTLpIS/LCtWZcSv/MqvUBUlf+7P/WmsFlppWZWUZcX5+SWmFX523VQslwvGY0lmOmfJsxylNPk4Y2NjinOSa8KJITemIYo1+SgliTURhkxrfvHnfpYIGBvpB/tTX/2cHLymxSrftKOR3rlax2xu5jS1aMf0nHvpKlUbqJqWarFg6aMAoYmmxKOEeCLJ37quPOVYouXxeCxNuH0BkNSVSAShVChqU11BUNnUKK1lbw2qinEiBhhyLuJth/0k+QnjE9ZhTQqf3bJaFTjrJOoh0GIDOChOYZalNE3brWHjcxxoUVGNlMxxrByjVckf+sJXhAdv5vy/futXOa8sr180/OEvT9GjTYxf02naO5ivuj4Vxn0ymvKzP/xHiKymcGfs7+8xnU1JxmMPSTjevneTnd099rZz4lgzGme8/to9ls0Fd+/vYmMj3YC05/+aiPfee+pbrdF3UsEXkjhFypTaWtJQVKOtCIQpfxAQUdUFdb3OYXfG0tKIMfaHZzD6TpSGiaJYZIcRvDc0E1FO6GmgMJUP20Kyt+PIR4LtWY2VaiucA2Oc19a23sPImU03sFaxWl1S1xWtDdREx2pZUJUVb7zxAKVSTk+PWK3mbExn0qYrG9M0DfOyYRVZJnFEpixppDqOrvN9N5XzHSGVHDyJs/yBn/5RFBFHpyc8uHuTremIVSl0uSyTxhrn80vOLy7QsaZuG7b37zCxiumi4t95fko8m8DmFpLy1phW4VzbVewppXrFQyeFZXEaU1tDrCMPZSGetpHEYtM03YZOkgTnRCN9lOY450u/dcRysUBHltF4JGOLxtiaumpA1cRJThKnEoV5IbTxeIZ1BkdL0ziatqI2K0bZNlVTUJYrYp3SmlIMDG2HdTvXEmkv+uQixuPa96kVwx4lirZyfPjwQ2aTEZPJjCiSIqFIa/77f/gPOTs74X/7F/8soyxjkuece3311aLE1o7GVjRtAxbauiGNc5bzgjzPiLTzsFhCliZkacpisRCWkvPtJZUiiTSbsw1wliRSTLKEP/9n/j0iDKlr0ZFj69Y1lLWYppS9Y4wvOGs9iyRBJf7Qd4ZMNxTLitptYZ2jajRFWVOZRiBJBUmWMZpNu+R427QslnOWxYI4Fu2m8WRMU5cY01JVpZAMkH3U0RBVYEB7gbMsYzadwmopNTBKYRthCLVtS9vWntElEXJd16g07pKlEtlLnqtpG2xrUM6xsSrYaE95sb2F8UqeOlgZb7DzXKrX29a8VIOBj0LFUdHkxRLzy79K8v4B8WSM+SNf4ejkmEel4+jsnK/evsXtWxOiJCNLE+I4+kS7+qkw7qN0TGrEoCVxzGiUsbW1yf7eNQLf++7tmygdkyYRWaLZ2h5z98E+C5vjIovxHq9C+mZqp0gSoSJ1uKwnqGiliG2ObaKOcieaLxDSjEmSkqXi4VorioJiKFwPCUTrQlXGJ/sETxNlxIDFWSdZf+1ijKlfSsBIh5VYKuDaCjSeUhXE+QWjDMyPUT7i2v6+aJFYyPOY8/Mzlssl1opnk+e5GO/LOXEcc21/n/l8zvnFBavViu3tXcbjCc73Wj0vK7I4ZZREjGi8aFIvneucsAdaY1AOKcmO4Nr2Fn/xz/4SaHjvo8dMJiOhouIwtiVKIra2N0nimGWxQs8L7G++x41fe4fkT/8MpROmueh29GOs/WYY5yMaH74GumiiJRqL45jGGpIoRjnZfMHAiyREQ/CglssVWZZJb9DIMRrlONf6sZUCtbathZ3TWpRKUEjForGGPJuxWCyJYoH16rpFKU3dVnL/rXiXTkPT1kRxhHWtx0klcaz8vUUaTGvI07zDX9M04fHjE/5v/8X/nX/rD/0if/SP/jGSRAqLLi8XGCN1DtY4Li4uiLW0/lsuF2iluLi8ZDYbk3qJYqU0WZqwv7fNqiiIIq+EqETqoqkrLyh2jclkJIbJtrxx9zr/wV/4U2yOUvLIMYkixtvbmKaiKQvpguXrOCIHzjdg0EohwopeS94zeuoWNrdnjE3Es9OVSA9EEUmakWaeATXohSDyDgmoQCZQtD7KmS8WxFpkMEajiWepgLOmN/IIXAhiN9M0FTZSFPv+swqnhMkVGG+BJip720hLQE9RtDhsgAxRbFcVf+Sbv8390zMuN2b8gx/7EQ62trqckvUiYVGUenlh3RWyCfMr9s5CAJcMylnq5SWro2M250uq5YrRySU/euce1aPnqHSEG41oyjmz6UZHFvik61Nh3OtaeN9JEmGaFbPRiK/98FfY3dnB0RLK8cExGafcunWNt758h0eXjwWb7sIgUFo8TBsp3vrKPeJEQnPlc2TSFDcmcxs0dUOejnyUL02bFU6626exUJ9sKz0U9aCAYUBBCpV+QiVraa1g4UmSMR6NSJIUh/UULEmw9nxi8dLH4ymT6QZlWQrLphZDI/ftF4yz1FY8hjSJmIxGXeEHGrJsRJIsMXaBcdINPs+lLVxbtdRFxWJ+wc7uDpPRhOOTU05PjymKJRubW6RZTpSI4bkoW+o0ZaQ1I9egneQrcL3kQKS9potPHGdxgnGGB7ev83/49/8M2XjCoxdHaNVgXYNz4hktLi9xTw5Zvf8YXS7IliUdUOaTpkrLBknjRDaYl8+ly32Inj7W0TSGLM9oK4E2okjarWVZRlVVvp5BUZUVWZZgXItTmrK44OTokNlsjLWSj0h0TF0Z2tqA0zQtGFuhSbFOeu6WZQ2RGKeirIl1TNs2YEVKWPmDKo5TsjQFIorVXCA6RHfoH/3yL3N0eMyf+pP/Lrdv3WRjY0O40GkKVmh3RVGJuFppmS8WtK2hNQ3OWpaLkkSDSxWbsxkRDh3DbDYhiRRpEvOiKrC2pakqqqJglGTMlyt2d/eYbW3grCFPNKZt+PO/9MfQKBINIDRgnea0pqFtDI0vXvL4oOj5Oyu5CQCniJySZtQ6pWnhbGkp21rYY86hs5RRmpOmmWeY1BRlSTEvaY0hTVKmk6mHGiSCrusVq+WcolwRRzFtKxRG67H1IF8MocH6sGqzd57E0RlTlBW1h2biWOatMXHXaEeoh444VuRZwniUUyhF1bZoa0id462DQ772vXd58PwF57MZ7771JsW1ayQe1w91AgpJ2CZRIlGZ6gsfjXU4/71EnRGNqShNSfzmdVonhIrjsuKH7t5kSoKJtrg+u87i4ox8o6B1lsvLy0+0q58K425Mw2J+xtb2Djs72/yFP/+/YTwZoRXUVSWouzeGf+qP/xG+++L7fHDyIa02KK8TjRP8yiJFLs45xtOxP8F9stBXCWoXo2wC9Dg4BPxW/NQ0S4VL6rW2Az4XkntDbFjr4C3IIsuynOl0Y63QII4jFouFeF4Drz1JEiaTMZtbU5KVYrGQEFQjnmkojgmfa4xhNpuSZXlP13NBRTAmlMKLEmCKQlNXNdZKkvHw4IA0H7G9tUXTVFTFioOiYra5yWQiTTmcjanqiqrWlElGrhpiJ/MQKR9REKr4PLbpxzGJI9JkgrERe1tTfunf/oO8fnMPZdrQMYKLdz+SpiNZTDOKfLUeUr7tPXaRgGjQcdzh53isNssSqqqVaAX9KwAAiMdJREFUquFYWgmmSYJppQ1glmUURemll6GpGkZ5zqJYUBQrkZIAsiylKkvcwhClJU1rGY0SUh0xzkZonWOoaFqB4mStGhQWq0SqQPjpI5xtSeKYyVikiU3bUq3mNHVDvVpRVTVpkpGlMYvLc46Pj6jKEtM0NGVNWTecrI5p6wrbtFRFwaOHD8nzERsbMxIdSfcmX408ymKKYoU1RoyJg+lkRFWscKZlOhmxvb3BZJIDhrZZsj1N+bP/7h8ljiMy7YhcA5ET9ceQhPZ1A5GGpjaSsPYGmmDMfcGCcZrKQesUVeVYFS2tUaAsKnIkScI4HREnckiXZUk5v6RuahyKKElI85zYGkxjfIVzvy+Vkn0zmUwkf2I9XdW0pGnq2Wd9CsZ56DVsr6Dlo9DkWcZoNJLKWp/LCPUl0uTFMRqPGWUZkY4oipKyaIhQ7M3n/Ow3f5ukbdlYrbBRxC9//Wt8+60HrDyxovWwjvNUX+XXhziDsm+D8J9Doz31GXw+zRjy0YytNzaY3EmxCn79vY8wH1X8yq//BqUdc+vB57lz4zYXi0uqqmRgRl55/UDGXSn1EJgjXI7WOfejSqkd4G8D94GHwL/nnDvzr/9PgL/sX/8fO+f+0Se/v6EqltjNDUKjAXyYbnGenicT93x+xIdnT2i0SIU6FUSb8KGh8x6Urwr0/G2hzEiXl5GaYpuIJEq6xSz1L7J5TNMwGU1oGoNpjUi1omgHwj8Q/i5k7nucbTqdMZ1OyfOcJJGuOahNdprGU8bkapqGi8tz0myEs47pdMxkknF2VvsEG4Ai9VhebWriOCVPR8RRgnM9N1w2QiyebWMl7E0S4dy71osWyWIuVwV1VYkc7Y0bHJ2cc35+zmKhBIsfT8jyCa0xrJqGUinSZEzqLGPXoC2gB+qCrk+oGu9xa6XZmkz58a9+GdPW1Abu7u2QjjKSErJb17EHBRdtxVgL+OMMJElMVUnVoLEO2zSkSYr1GuFJklCVFVpLiN22NVmW0Ta1MCeIKVYFaZZijBwASSp0wSTJBObzwGxVVSRa+oZWdS0Vyk5LlbOuaesCnWiy0Zgsn6CdYmtjg8bWvuZAciBpnHN5fk6aeUy4qiGBhT2HCCbZiLZufbLe11M4S1kWfPToEdPJDKUT9vZ2JRqzLVkSs7u9RZzE5FkiDLHGoJyiKleM0ylJHFFXDeN8TFNXVMWSySinWC7Ymc34D/78nyeNNVksaymmZpSK1EXoxyQ8/laotl7DPvVJDoEFY5SOME1DaxWl1dRNS9NayqphVQs8GSUR2Tgn8zCitY6qKjg9P/Xa7YrRaESeidhXYy3LckldVThnSaOILEs66iAEerAYv7BPkyRFJ5JMDQnz8BzBuHsb1FNqnWgHpZHsj9VKCpmcC4VwQpiYz+c0VcUozxmVFW88fsT9x0+4++KA+wcHPN7d4bfeuM937tzhfHMD17bYrlObQG5CUxZWTmDUScGR8XmzFkfU3W9o3JFkKZQV0Ytvw0rYQa+3Ld88meJIOV8VHJ+eMJvm0iGqs4kff/3/4rn/vHPuePD9XwP+J+fc31BK/TX//V9VSn0e+DPAF5AG2f+jUupt9wl9VK1tmYymXdgfvE/xSgHPmHn/+ft869Hv0ahGEo1Iok952lGY3MgKXCBJUtc9qDOOxGlSLdSyLM+6exgyUgCUU8zny3CmCM44LGbxIZZMkHgmW9sz9q/tsbW5TZ6Pe8OOYPliU/pns9by+MkjglOrTMTO7jZF0XvavVCQ7ni4o9G4N+6uV/eThhQpFtsJVpnEUFH14+n5R9ZYLi7OaduaGzducLlYcXxyyPPnz1Eq4tat24xGI+I4oW0biqqmJcJmOTmGkROaWK8OabrIwiifjlYRiadtpsCf/KO/iI4iLv7bXydPIh7lis+UERetIYkzjHU0Te1DZk9BjQWGSZKExhqfUE0oylpyJ3EsjRvyHOcci9VSjL0vSMqyjKqu+zGsm66Ap21bprOp7xQkRn5VFGSZaMRoxCDjNUSK1YoIqSBM8hSrYba5jQaaumU0mWCNpXE1kY6xJgjB+fm3nkfh2xROplNmkzFTX03c1CtM04AT/L5taspiRZvnVOUKZyxJFBNHisXikul4jIssf/yP/EEiBblqiW3FxjjGNsZXNnq5A6ehldoO78lIO0nlAlSM8ckVqzzsohPqVnO+aFiWNWVd+6plyLOcNJ+yPUtwXvJgVRSUxdInjy1ZNmY8mxG1IuFR1g3L5VIOXX9f1hqRL1ZK6IB+f2gdY62iKAvqtka6GqVCYY5E6kPyKtrn5eRSqvtnv5ehWytpklCgqJtG5lhrv4fANg13jk+5t1zy9nvvEy+WXGQpH17b5Vc+9zbv7++xyIQmajzkh1ICY0W6swmiztnbh1AM1xM46X/uXJdX00lMU62oT2rKxuEih9LiGKVJQprGVHWBc8YXSP0vx5b5E8DP+X//F0hv1b/qf/5fOucq4EOl1HvAjwH/4mPfycF0ttFpqQu3WTaDcoKxPTx4wrc+/NfUVLhQwONC6CUnvXbSLR7EUe9Jj6CMIjKKNJrhmhic8I+t93otXsDJOWxrSbMR7dkS57sjhYNHKFq+I3kSed1xeHFwyld++Evs7Gx3hrsolhTFisVySVmUlFVJUUoTgLKU/+6/9jrb2zvComgbmhJG2RjjGoqi8IncjCgyNE1LnkuYq5TuFoo8rPJ83xEOQxynQEySjoiSUtp3EcbEj5FzrFYrFh89ZHN7h+v71zg6OuLk9IwPPviAGzdusLMj/VbjWNPULYuyooygTmNSp0itlwZQUkFrPdVPRw6tLKY1UrKtpAxeK80cxcaNPbZyxcbSUGvxBp3SEGkvfZwCRvIHaeY7YTniKKYsCikMMdKcfDQe0dYNTduS+763CmnVVpbisUdxysXFBcYarNHM5yuSJKduLY21jMZTVkWJaRtUkrNYLJlOxsRx0jsCzpHnKUkyo65rNJBEkUR3PgxvTI1DStmFFWUxztFaSJwVdUVnhGqqlffovOyBsVjPu2+tpTU1q9WCqlqQJwm/+PM/jrGOjVGMMpZM1YwzDdnIC+EZGuuoW0fT4vnjGktE6N6tPP1OA2MruZMoAqM0jZOiobp21MZS1g21aSUfkGWkWcL2dCS4sbFUVc3lckFRlh3ePBpljMYTQLotnZ4c+d6nap3R4izK2YFhT4ijDIU0uxE2WEa8TCi8MJ/WMs6N0j5K9QV9KuocQlnX3j4MLoUiTTPytqVMStq2IY0jsjTGtKCM4XNPnvILv/MdVlvbuCjmnfuv8Zt3b3G0MaH1xYidgJh/jtDtru8m5Ts5pSlRknomkPO9AhzO5/C0k2SzN38orUmnU+xki+lmQ1lXOAf/6p33OV0Z7r/+OrPRiOeP3qctFtx/401u3rjzsSYVfnDj7oD/QUnm8P/snPubwHXn3HP/QM+VUtf8a28D/3Lwt0/8z9YHW6m/AvwVgJ2tqeBU3lqFxgvOe8xPT5/y2x/+JoVbSTjjExLSQb5vmiyYu4RqsQ264v7A8B3vU6bUtfE8ZdedvrjQu9B5xo7I6sY+1MJPgLQMSyQDn4mn/O1vf4fDo0Na0/LOO+/w3nvvc3l5yXy+YLVaslqtqKoSY6wUxri+MGl397v8yT/573Lj5h7GKOpaOL5Rx+Om80TzXJJuw/AzeCeB0yuNLhKRCfYRRhRLK78u6Rxe7wfOtJaT42PSNOXatWuMxhOePX/BwcEBl5eX7O/vSbIujYm0om5qLsuWOIoYRykpjpGtPfYZGETOG60gaCXeYGsM87jl1o0trt/Zovy954ycJYkTCt/4QOkI07ZEaYSOFIvlUppSaE1ZFqRp2mGcWZZRV+JNjkcjVsUKrRRZllIUBWmaEycJ5xdz3n33XWkOEWWA4snTZ1zb22EyGVFWFVVVM52MBC9vG2HTNL4Bg9YsfIWrUoqiKKTPaZJRtCVxLElf0zYYI4l5UWSMOrVF54KyoE/AG+sFziQ5WBbCr/+5n/4Jbu3vYsslk1gMYWob3ry9h9YxKIP22vJ47+/o+JynB8e0aU6UjIjTEYmXo9aud3OU9pW9SrFsQ+WvJCpbqyjrBmfkkMzzMeM0JoojmlZ02S8uzrtinCiOSbMR0+mEuq5FbXO17KKm4MmG6PXqenWefBD7pt+Bsx7yW2GvRX4PdhXgPiEqTaX9erad8AR0DXBevrI0I89yqroiiTXbZcXb3/4ONz74kMRanr12j3d+7Me52JzxdDnn8vJCEseh7Z0ThyWsdRPYN4NnC/culGZ/T0p5jL3Pk4HzdMmUIIYXbewTN6dEtSYnx+oVERU3b97CNTUf/N7vkLiKnemIe3c/98pnDNcPatx/0jn3zBvw/49S6p1PeK16xc9eGml/QPxNgPt3b7gwMCH5YH0rutPFGb/17jeYtxeE0nNJuDm0U6HHjbA2wg1Ykc3F9SqEWilUPAGX4WxNFAVIRrzyYaZdRYnginXViVJlWSYsEa/Ilo9ytrY3WK5Kvve9d3njjdcoi4L/5r/5f3N0dCyFT43xYlgeQlIKreW9JHSFgxdH/ON//E/5pV/6E+R5jtYVUSRd55MooW5qkkhYAc4qKUV3qivWuHrFcYxzScfqWa1WvrGIDwxdL3AbDlChcQoX++nTp4wnU15//XWOj49ZrVY8fvyY2XTM9es3GI1HRIkoX9ZVxdy0jJIYozNSbXFGoZ1U10qOVTQ2wudZB+5zr7Ha2+BodcnhNccPRaARnv0ojiibmihPmdcG1wpe2tQ1ZSnJZMEuLaN8RFWWKCVCTstiBToiSWKWhWio6Dji4vKC4+NT/tbf/q9wkSLLRjR1zb/8jV/nf/dX/hJvvfkGxWJOPhqB0qyKJYkX22psy9h3wZrPLxiPxtRNzWq5YDzOqauCqq5Ik4RytWJ+eYmjRTmLaRvfos2Pv1LEsWKU5ozTjCyJUdpg21oig9gyy0b84T/wk0QtKIyPZgHVSn5DiW6RQfTTTWV49PQFq1XF3s4uddPyG7/7Ozx5ccR4Y4PdnW32dna4fu06k6k0lbk4veDJ06ccnRz7g1eqMq9fv+H7/W4L1bOpWZVLxuNJj0m3Dflo5IXVpBGNOC5mDQIOmHfAmrWOO09V9c1dO7KEOFG9DHFwvJI46WoWuvfWeg2Xh/6zFYookVaD4kitd6PSKiKLYr7y9Bm3L+fcuZwzvrjgZDrh+w9e5/nnPks72wZnSapS5J0R6DeOY9k/thde06gOCwqCd+DzccMBccKeUUqiKK1jNqab5HlOUSw9FVuTjsaw2UBjyI30aI6I2NzcYDSb8BM/+TXi+SlqOsXxPwMV0jn3zH89VEr9XQRmOVBK3fRe+03g0L/8CXB38Od3gGef9P5ai7i9s31lIg7OFnO+8Xu/wWV5IYVFvoUVNiSEhPxvEQ/GOjHqkZfs9b6KLCELGVOsFWOhVa+CJ4UWbef9jkYj0iQlzVLSTKoQg45D4I9vbm6wvbvJycmHZOmIu/fu8J3vfJenT1+siepLJWDP9plOxaOULioK01reeecdvvWt1/mJn/g64/GILEuIk4T5ZSFYvIpFyyUGpbSvcJNBGmL44f5E81l4vKvlkroq+01DfyTIfUnZvvWKeNZanj9/Tj664MaNG2xubnJ4eMDF+QXL5ZLdvT12dnY6eeO2aVhWNYXW5OmIJI4wtiF31jeNCBV9fU5k784NjFLsbW2wu7Plf2e7atxMOVzbMFaaKE45W8yxrfX0xtYzKKZUZYm1jjxPqKoKnHhMq6IQ3Y84ZrlaESeCpddNQ1NZVkUpCToFZ+fiiYL/2+Wqm7vlaoltxSDUdeW1RpxvaG2JtKauK+paJAHKYsVqsSDNE5xpRCLamI5JVFY1yWTMv/WH/gBYw/YkQTklVMamZBRJtjV2ocJSwDQFHr8XF9yaFqs05xeXvP/BM3Z39nj9/k1M27A4POGjJ4+5WFWcXF7y8NFHOGB7e5uvf/3rfPjBhzx8+CGLxYK2bdja2ibPhSX19NkTbt68yVe/+lUAmqZiY2PW5SuMnXB2XnF+cdExTWQNhpX18Tk+UQUNbBY/376GRfvkY0guymPKSk2ztItcQfIkcZSSpjn43gtSXOdIUkcUO5HVbbykhf98ZS3j5ZI3vvc9stMTJifHTIxhsbfH73z1q3y4vUkzzhnlOdoJdTpEDDjTa8b4n+OTuai+yA76yHVY4Rp+rXUiHHcVMZvNSJKUs9Mz0jRG+4i8bWuifETjWkaJ5eb1G8xXz3jz3l2MXuEmMfrCMN6aUqm+z++rrt/XuCulJoB2zs39v38R+D8Bfx/4C8Df8F//nv+Tvw/8LaXUf4okVN8CvvFJnxF5LQv8QoaIRbnkN779q5wuD0Qp0Zf5409K7QsnFEqUHoOD4w20DYbP68rELiNyY1ZlzaTToZCuTlJNrHx4GDOdjGnbxutVSPl5kqTSGDpKGOUZGxsT4iQiihOiBN544wH/+X/+/2RVFN1zOWfRGkIB1Hg8YTQaYe2Mra1ttra2uLicozUsV0ta05LnGQcH58znS0ajmVR6yponiYcNFuTqpYLl0jryC19em6QpVVnQVXLide4J3lVoCmL70Nc5qrLg8aOPuHbtGm++8QaHhwecnp5ydHhIWRTs7O4yHo+7HERTG4qqoFKKUZqBFgEsTSPz5EKLORhpiJUmVuCQ1gxOIRCaA3zlgnaAa9gZZ1R1y6pumUxnoC3L+RypPkw7vDfPc0oP14xGMxbzOTiIk4wXR8ee8hYWtnzUi4MD3n77TUajEVVVUjc1k3FIcDZkacZqJZ2bRnlOVdUsltJgYrlcslyupMmGdhRLyaNITYRFlEgM2sE0dqTakZmS2e6EWIFpWrBIdKfDfBqRk/C0Wu3pfCg/mo3lxemK4/MLLs8uuHfnLtbW/Ma3vsWHHz3m4OiYRWMwg/DfArt7e3zv+9/n+9//vkS0PpG4XBaUhU+4A2VR8dabbzObzdjc2CJJYpq6FjXFYin0zbb1dN4+2pZ7FIdr3YMPJIKh8e+VDcGRRpK7Up4Z55TrcmHCFY+7lnhRFDHKpkRR4iNhSFPp4maMpS4czkY4J55z0tRsnJ1y+6OPuPnoETeePOHDNx/w8MEbXNy+w+UNcTRUW5MqhSbytypRehTHqAClesJAQASko5ruDbl/xkhHRCowefoanCjSbG9v0zQtq+WCi6YRbn2UCsTTWup0jAPaRYHLFHv7I+aXBbtxxT9bLEi2p3zltbu8W5ekhw9fsqXD6wfx3K8Df9cbkBj4W865X1ZK/Svg7yil/jLwCPhTMnnuO0qpvwP8HqKa9B9+ElNGBkW0tp3n2RarBb/xu7/O4fwFxtUorNChVK/ZYIPHoDwTATHwvlDVs0I81q4USTzD2RgrcX6n3ayVYjTKaRppPpukCZNxjrXGG1Ylxj1NfNWfIo58P0SEQvdTP/U1nj55xqNHHzEa50wmEyaTCXu7O1zb3+fWrVtsbGxIFWoszRym06moTPpS64CRt03J3v4uaZpjrQaMN+iGKB6GtHKpK9ZePIeo8+SzLOPSJ+1CswtrZXSGnpfWfQgNomljneXw4AXzywtu3brFeDzm5OSEqqx48eIF4/GYvb09DyclRE1MXRcsq5IqiSmjmMQ5EmuIlSXW+IPKSVLcifSxscbDcf1BJY0JpEBFOYuKWtJMgV1RNYZJltNYR1lWXTRV1lJGPpvOOvGwcT7GGMuLw0MvQyHrJ6y8Zy8OiLMMHQmen3gsN2jUJEnEYrmgaWqmE6EcNk3NZDRGOUtZVhhjyWMZszSJGGkLpmWSRUTaYUxJniiUMmhr0FYohuFgDdW5xjdlVi7Mie6xchVhrWNVVXzn3Q949uKINM/4re+9y9nZmcg+AHEojPGUQgdsb24ym8345je/2Y2LaPRoyqomT1OSJGFra4s7d+6QJCnL5YrHjx+RZxn3X38N5/egGqyzMI4ifGZRGoEOvZPQ514GETm6G3ulxOClSUKkFa1pUURrnrD2MJtqxU5Y7Yu9kM+yzlIUOmQlAQfGMrs858aTJ7zx/e/TxjFnu7sspjP+xc//PO9+5jMsbUOajnwRlPONVghMDJw37nEco5q+962LrDTa0bqrmwnPanHoKCKJpedsWOmK4M0LnHl2dkqQ+Y0T2f/S4Kfin39wxOuvvcnWdkSV1/zInS/y1mu3yFLL89GUR6R8I7I004x7l6tPMqu/v3F3zn0AfPkVPz8BfuFj/uavA3/993vvwV/4BakpmoJf+71/zrPThxgMKNst+NAUQS4V8hQ4JdoP+CSOw/kqOifdlqKEhA3ayhBHGnxbu5BwTNMUpaRTfZqmTCYzFJrReCzJoK7RhAMHWZ5Jl6a6YTrJufHGTb71r7/NH/w3f4Z79+6xu7fLxsYmk8mENE28JIFfeMqued9SMk2XGIqimMlkBkScn1+gHMSR6xpZhM3yg16B4SMNAHpaVjd+WuhwiYqEV269rID3/KUTkuPF82fMpjPu3r7NfLHg/PKc5WJBVZZsbm6xublFkibEyVRE1uqCoioZZzmZTsg0oCypLxVWTvgF1uGV91wXSQTFPGNNtznwXW3iWJPHGoMYyigRzeymKYmUYsMn9+q6YpRlKKAsC46Pj/BBXZ+pcIrLywVpmlIWBdZAmsc0rTTFmM1m3cbTSg5D11oSHImyaNuQKkuSKNK2JlWWaaYkFtEOqyIvWKclseYcxolrohPXGQfnfXxlDMFjiZTXNXcAEbWLKMqWk4Vh/9br5LN9js9O2TARlgx7diosIeVL7514girSvP7GG7z//vsyit6hieOYjdmEra1tbt+6xfVr15hMJp4JIwdAmqT+oFvReFmGYMhDFB0gCOclq8Pi7sgKAXLrEql2zejHUdzpojtrcSq0OexWMEmSE0WSUwuFa5LcdJ0zEjbTaLnkq//yX7J5esb57i4Xm1uc7e7wwWc+Q52mPvFpia3/Oxegmx5GCfcZ6UiUO6uaxrXEPuHbNtL4QyO5gsgnflsrVNXUy5RYZ7ocnXMiOxzaHzor50hobh4OQJtfY6VmzPav045irue73N2ZYtHsnR3ysJqzilKccWxu3/jEvf+pqFCVy2Fbw2++81s8OfsIEtFb8OvUh6iil4FSnUaEpeeQKt/OzABEmsif9Npl0EZUdcl4NPKYnizCLEuZTCYYkwqMkaQkiSQ8Q+gum9PhtQP4/vffI9Itt1+7z717t0mTiJ/+6a8PPI7+joLhDj8PCcyQYYeAfbf+2cSbzUc5o7qkXDZs7zkaYzg/jnG2cz8/9hp680qJtoZpG4zz5dF+nDrD6Q2P9LNwvmzedntGIYns+eUFVVlw4+ZNtne2ePHiBWVZc3Z6yqpYsb29w3Q6Q+mINBsRxcLVtg7aNKVJEhpnyJ0CTNe4I01EPtfYIKrmhdSgU9qEXlfG+QKvceT52FSMkhjrHHU1J2lbskgTRZayaYkw1L5XZ590kKdfLpZUl3OcteRJQqalBVsea1LtUKYixRBhiJuSWGtm05wk1kCDTpAWjqbBhKHs/i/RkPWsIeVhpyhKOrZFWIvar1tjLEQJRiXU1rGsJQldm5pVUdE6zWgyJZ9M2b1+HZAuUovlnPOzU6GynpxwcXFBUzfcu/capS8K+7f+0B/i2bNnNE3DjRs32dreFazbr8G2bX2i33XrSHI4IaoLLJShURVDLZLLUigY1nSo+O4bqbsuqQp9By3/y05ye/j5SgmFsUv8D/rACuBu2by4YLRYMLu44OaTJ9x69Iiz3T2+9/nPc7q3J+QKT58e5r8EsjJrUUi3U/3nxZFICIRuWUQRdjSiMSKQFkcR43wkh1PbSPGgZ/HoSBEngWlkujyNaCYl4iQOuPBKx/zIv/E1lDVM8oxER2jrcDoiSSe8rs44NSsKW1IZxwOV8uufYAc+FcbdOceqrfjtd/417z9+B6sbwdnBY42yZlQSuDGyCowCp8XwBHvnFB4LU6AkRB2ZTW/x+xZVkZZkyWg0JkkkOaiwPqtPl7SU6lZ/K04247e+9dv8/B/4Onfu3AEl0q3Dbd39gfyVzxD1IXj3vldeGVg1DgvKCQ68bDg+cExmGVHc0lR9haUK76+GmhpDrykkpyIP1fhF60PiANHgE9POywnLB4Q6Ar/4/KeVVc1Hjx6xubXF3bv3uLi49FBNzeHBMfP5ip3tHfJRjosNJBlNVVNWNUXVkKcpdRSTK4UxFZEx0PpsiwPrDEq5LsnehfVBSbNtehxXKa+CqVHaohWMEg1JCk6eJ04V4zTnz/2JP8o3fvO3+NfvvMflfM54NGZra4MHr91lFimyiRRBRUoaM7tYo6w0Mt9KI2waxLEg1oAzPunnaJztxlLshOdq4xVKA3XX66EoNEKXlmR2Y6G1EbWKabSiqixlW0iXKhfErMQwqrBqtDBCnLWkScTe7i43b1zn85//AnVdcXZ2ztHREbONDT786APeeuttlNI8ePAmeS7GsmkMy8WC2vePDf8553yv0CV37txmS20RxxnQEMctbWiagnjZqttzFu16rF88bNcV44W/CWvUWkWapCgdA1HnaIQobriHwrfDvqHbJyfcffddZhcXzOaXNHFCk8T86s//PM/v3qeJvEZj99n9+/WOlWUNB1q7lGD+ibQuTJIU55+lbVtpfZmkvtm78O3TOGU0GrO1uYkxthPyC8WZQn3MSZJc+k0M9qwYfyXRbuRJ3kqhVIIz8NYHz3nj/JRlJgfj9MOLj7lvP1af+Nv/P16/+97v8M7T36HVjcAgTgWGok8qhZPWeT8n/EyJR+0L7wT5EI8QHJHL0Sb37cXSzhuQjRaTprkUS3kDpz1VybRmzQCLnZF7aNqWyDNoWl9tx5qx7tUiA9Z4FRv/OGglsAictVjXorWjqlqaKmVrK+HspKKpuy2O0tIQYejtDN9fvmoZMxeSXfJE/fNpwSsdvsMThKRmMOwyzBJSR2guLi6pqppr165x+/YdDg/FwC/mS4pVydbWFltbG2gdEacpKtKSsGwb6haKOCJSCZmKGGkvWmVrPHEYh+6qJ8XYx68ctxCih/URjKz10sfWCpT31r0bfO7+H+OPXy45Oz9jc3OT6XQiYbUGJzWTvvuRHOgq0sL4sXJwOOebU1uZY629/IPy+kRK1p11Iuur/cHetsKTFn2clIaYwkjeZ9XUFI2hqI2XPQjerCSP2sYM2FH4s1yTxqHpuuk458p7gVGcsLG5yXgiieGvbv0wCqnabpqma5PYNgN64WDdVFXF9773PZbLJVmecf3GDRQxSlnfVajyErZgreSKlA1tCV03L6KT33bvv742w8EQEcWZzwF5sxutr+HhFRnD1ukpW4eH3PngA9Kq5OmdW7z/9htEjeHF7dvUnlAQHAO3lmfx+2zt21ftRZ/AjROSJKO1hjROsVra8DWm9b1n8VXDUjm8ubHJ7s4O8/mcoigGe8yDP1qT5+N+nXrtKqUUeRSRKlC2wRkNOkb4WYq4bRh/611G7z9hpSyJikjtxx1Kcn0qjHtRFfzeo9+lUgaiXggLyZFiFaLu6MNe6z0hHCJLG+AZG5TWpHBDO8jYBCJaYxhlUdd7VEeaONZEkRJPGYKbO6B59Vfw4ENxxmQsbfICRevq8ujx8f576bkqVMvh63qvRiKGZ0+fc3BwwnKx4rW7t4W3vSxxLsUavbboQ7OQ9c9d/94N7s4zi/Eu0pp3HJZgXXvZW687rZTG+arESEuHKZyVhiPPnjEaTbhz5zar1YoXBwc0Tc3p6RHLxTmbm5vMppJ7CF2q2ralqWuKpqGIEook8UY2Joo1iTNEShErvBZL60GukFNBNHq8tKpVw8dR/UHrgtyC5E1wir3tMXs7E/mpijAueKwyV6a1XVQkbfpkA8a+GYuO+nXpkE5YjW/yoJXGGtNxup3vJaBD8tc6qsJQtqIe2rStF7qLiKKUaTYizzKiKOLRow/I85Gs5SjABv4J10TrtKcQCmYtNQ1e85/eWK+WS1bLpTCLfM2FVMn2kElY8y9ePAdgd2+Xg4MDfugLX+yiPe2bgYQ1qz1jpHcmekPeab+o0Mw57LOoi77C4Wy8VooeHALDrwqYXl7y5W98g+2jI+azGQc3bvD09m1Od7e7YsUOSx2ucXmTsCO6+/z9ruAwxHFM1IgBd1oTRzGjfITDMcpysiQlH43JR2MR5zs6EvmTftN18xc0YZyHoYRqqcjSiCxW2LLqhPM6u6A1TVGhzuZkxtFgyIEU/eob99enwrgvqyWNrQCLVQJJoCwOTaxjlLXY1ooHZXwiKon7h1cQOWm4gQVtQTuDVgka0TQPmW9pdCEJ1TRLhAaJkQjAeRpTSHLQY3M9bC5eknBvG/+R/UEwNNTDbjAoYQY4Jwk6YwxNU1OWK5arFXXVcuPGLRaLgnff/ZDTk0vOzy+4tr9HnAg9a7kwHpIaWnO5p4+93NB7GRw69Jl8j/JgHRgnXXGsFYOmECgCZVFaIqRQLh+wysViwcOHH3Dnzm3eePA6H374kKYuaaqWo8MVF2cx09kWk+mmKF7qmHyckFg5ROu2pqhKobEqKdBKk5QsjtERJFaocJGTnIeMqyJCSSM3OW1QVu7VanH+nLPE4Ofb6354yAtE0yg0aejm2WOlwfjIuvBhtd+YxouPOatpFbROi8HV8jmVUTRGALbGOsq2omobrMdedZSQj8dM8pFUOqdy6FVVRVlXFKsVx8dHIrSVT1BKd8nOjs1hLa21XYW2xSeg/URbY5jP51xcXHQdxMKlPD9YcgKWdYfAcXF5zq1bd6jr0q8NJXgz4hAVZYVuJbE6LDzCj3W4xMuXnIDkWKS4TXnZBaUi7zwEr98fDoNIJbKW8WrFqCy5+fQpSVny/O5dHr/+Oof7+zRDzHptW3QY5FWn/cr2GIzLlWRw+J3WWpLLWrTjszSTxLKC6WTC/t4+aSqSysvl8krOKxxQfmz8XnTWYayVblZZThxF2LalY/yAhBfOieZPrKkBgxbFWOT7T7o+FcYdRL8c7xkQJthCFIvBprVEPrxyANagvecmobGBSHsMXsLgRE1JVEZhSiLfW5PO21CkadItAjlXRRUxy0ecnZ3Ina3BLeLjj/KRCFq1oiESDpjOU/fOgzHSm7FYFVxeXlCWFWVZ8fDhQ54+e8ZivsBayPOMnd1tfuEX/gDvvfeQR48ek+cpP/TFt9jf36Esq84oDzHLV12q8xLCyNIdVMPtvRZrONcxaQJDRvucRJCC0Ep1HOQwZm0rnYVQirqqePjwI0aTKdeuXaNtKs5PjqkbabKwKg6JTi+YTCdszDbY2Nwg1anHY2PhTrd4jW2LaVsWTQhKxYO3HqPO0hQdxUKtVKBd5J/I+F6wqpNviIKRRpLzyo+H7tZSgtJ94tv5zeSc9WwWyVW01tAqB07gg8YYlIswSmFUhKkNTdvgnKFuRQcIjTcK0pgim8xEKdCKE2E9tj2/uGCxmFMWBa0xJJ7aqZXm4cOHvPPO99jc2OD69evcvHmbne0d0jhBITWKYe4gyAkYLi4uREvHhKYVAxjuFdCWtCZsWBUF8/kcc10imMQrjQqBSbzvyCcSjcXnbHwNxcCoaa2IY40iom0gSTQ7uxvMNkecn604PlyhlMgO9PtrsI6tJW5bbjx7xtb5OZvn5yxnM7754z/Oxe6udwJsJyPtl/HaO3VO+5X3/nhI1K39OxhmraWbU4hy4jghTXOyPOf27Tvg4OT0lKY1vCrqCPMylAtXSpHGijxJUFgxWU5sl1MWE/arlkiNNMHlqYCHaYJrnHTn/YQI5NNh3BUQy0IRPruUK6uWvngpkhdarXBaOh0pp7rSdhxo56SbulI4F6HMjNa2mLZhsrEhL3Ou28hpYIV4HxYl7fDqWvpgvoTTAU5FzKYbmNry/Y/e53I+J46lPPjmresY0/K73/4ODz/6iKOjE87PLzw74QaT8YjZdI+N6RY3vnKbuqlZrpacnp0wmYwZj0d8/vOf4bOffYuiKDg6POTxk4/Y37sxWKi9Fy7XUCTJ4ZzukqcgHkIoDxdvrS/57q+efSCMhP4Q6bwQX3LtdCQKg87rawQvVoFzmsvLS05PT7l35zZvvPkWh4cHAMw2tplMNzx7xrIqlpTlinK1IosT0iwjSjOyUc54OgblsK3cY+iYgxF+fNUYtHEUthEvXnkvtAVnhGssh5NUekI/k62H3JIkEdaCkpyFcYYQwrRtg7XS2s4pUafse9eKx9V4GdsoinwRnufmxxGjBCYT3WkQBe/UWkmwFUtRTlx6Y66U9Cvd3N4gSRIePXqEMxanYTKeMJtMOT465vmz53zn29/hK1/+Cg/eeEMqa7XyydW+EXNRFB1rZnNzs6PidbOt+3vTWlNVUmVrbMNyuWC1WnF8fMDm5ia7u3voyO8RJxtRqQiFaMJYlFcwbfsDNopJ44Q00WTjhL29nChuuThf8OG751SVMKKSJOvWYjgckqpm8/KM+x9+iFOKJklYTSZczmac7e4y39pi0P10LVIO369dCm8Pfn8cZmjQ+9xcDxOFn8fxjJ2dPYwxnJ72B2inQTUw6FfHHaTaWMeOPI1QhEbeTtJeyrcqVFryaZHG+cg0yhJs5MjuXaN9/znpjevw/JCPuz4dxt1JAQhKCybunMAxOL9wxZi3znoqoITBOJEntcqReL6q1eLOjeyMxE2p2wpU6Bgkkxd5b0SpYasqh9awKgqqqvCh+qsxrfuv3eHJ8yO+890nlFXJbJbzzrsP+cqXPst4kvH973+fNMt58OAtsnSMUorbt2/z3e9+h6YxXFwU/Mqv/BNWqxWT6Zj7r7/GZz/zFsdHR1xczPnww4d8/3vvcXh4xC/8ws+yt3u9u0f3kh8yrAhc/51QRS0heWoDqyO8y2C9C37bY5EvJ4Bfged7wxqM3Gg0Js1HZFnGztYmSRJz585rLBZLlqsFp2enlGVJ07Q0TY3W4h0H3Xo8xp8msagQJtmaCmeaJp2n192TT0q11mAjaXIsicuw8U3HUQ/NTwKlzmlN4wKVVnm9EEvbiuEOhSpKKyIidBI42NZjy5EvTPOqpH7MgrEsioLVasXGxoZEhFnGcrliVZQ0dUWep4wmoktel6UIzNU1ZVkSKc3jJw/J0hE/93M/x3K55MWL58znc27cvM5iPmexWFDbthurUCTz/PlzvvWtb7Farbh+/TpvvfVWN0ehgUxd18zn8y5nFArrHj582Mln1HXJjRvXu56/oQm8wC2qW25S6a2wxhDFEbNpzo2bO4xGGU1dc3x0wcnZJa1pvR+iQK3j69pY4rbhtYcPufP4EaOi4NFrr3G6s8OzW7dw3jDiD+eeDhnWwsvG++N+f5V0MLwCRHd1/Ye/k/Z+Izmcl8vu4O7lv/VL79c18fDvm8ZeidPntbqoyu8x6xyJUtKTOYo62LTJY6o0Ih7FNDRsfOEteP6dl+4zXJ8K4+6g617iHGA8x1nrwE4Vrqp1/Wvwp6sRUVOSCIuArZECRY5toK5bRqORPwx8AYWCNMsGk9gPvvQpFanWYU41gCJaKdIsJx9lnF0c8eLFc3Z3N9na3qCoC27c2ufW7Tv82q/+c58xF5rZbDbjzTcfkOeiaPjjP/5j7O3tMZvNiLysQLEq+If/3X/P06cvANjb2+X+a6+9hIm+fK03yr26aJNU2AjGWKIY4jQmjWLapvLNAwIlsq+MXPfc+9xBpKTQazyZsLOzy2g0FnXGVNgby1XJfDHnxcELlosFTSM4c+t50nJIyHtGkUMnocm4AyMecVM65peXoAI05DtSxSl5Pup0SMQDjX2jYPHY81FOPsrXtHKMaTt4yVqptIz94S6ceoGcjDE4QjJd+cbYvRGxztE2DXUj5frW1iyXS9pGEtCtsV4Bsu6S8kppDg9jRiMRXrtz5zav37/HxfkZZV1zeHzCwcEhTVN3XuhoPOb05JgXB8+4PJ/zpS99gc985k3yPOHJkycUxZJYR6RZTKJinFJe1kKEvN555x3yPGc8HnN0dMSDBw8Yj8XJWK1WXcP3NE3Z2trqdIJQlh/5kR/hjTfeoGkadne3pT+Ahz+Nf75gAK0LEJ1QPrM0486dPUaTmNWqYXliwGjqWuCZOHKSZCZBa0eUJChrScuSz3zve0Sm4XRnj4f373O2s8PZ9na/2q8Y6o+DUIZ4+RCqCv8O3Y/6tf3qPXMVwgrjWdc1FxcXXfPsqwY9rNXh33ZJa2NIYsBY2ha6W/EOBm4gJ6IDW00gRGUV5SimoWU2HdMohdrZ5JOuT4VxB3y3JduHI3K4y8NZ5U98jYuVQDhBx0L50mzEr9WtAZehVI4xFc43T+6sM+KlSVVqXyIN0tIrjqMuAflyiBeYzY6d3Ql/9N/+OdqmQWnp5OIsnJ2dcfPGDX7u5362iwryPGdra4vJZNp937Yt3/zmv+add75HWRV89rNvcfPmTX7mZ36Gp0+fc+vWTR48eMOHvNYvkE8OLT+ODhkW2yjPMaalaaSnauwThz2e38NWzgXaVk6WZYxGY8bjMbPpjCRNQQst8PjklPnlJXVVdXi5GFDXsTWcE32XPBsRNMzrpsKYmrZuAClWUoj3Z3wdQpT0rBAQeGa1WvnkaF9dPBrlzGYT0buZTFEo5pdz6qbm5OSUs7MzAuOiqqURSppm3L17h43ZJs75JtTzOcvlQmA52w4OI+uTguF76BdUgMJADlk3SKTL64xpmc8vWS4vefz4IRuzDUb5iPF0yu3bd7h27Rrn5+e+wfmKpq7Z3tnha1/7cc5ORFt/sRDP/dd//dfZ2Njki1/4PFtbW7StYTqbkSQxq5XAMaenp7zx4A0RNPMt7haLhV/jcRdFJN0BF+5UjN/u7q6X5nCdYY98M4ooimnblLouuwSxc5bYN2x+9uQUHWnGUxHG0j6PJj6V7tSjtNakTvHgo4/YWawYr1bMN6Y8uXsbGyWvzA8MPfar3vXV14WfvQoWuQqzDD/rKmUzeOtZljGfz6l956Wwt80gkggFX6+CZcQRES0cZxEVTNXH4cFjRyvf9Ft1VMkgm6E/c4fmd75P5BQtmlVV8UnXp8a4i3XWREniwxAf+jpN4wwuVp2BU643siCLRlkRmlJNxDjeJSWlaEUdcMgiiGKRjIzjqFu4Q1gjcODdOhFWfuuxQWMtH33wEd/4zW9hsd57jNncnPHaa/fQOmZre4Pz80uKVUE+kvZrh4enXSh3cXHBf/ff/fdUlXQJevjhE77+9a/xhS98jjt37vDs2XP+yT/5ZxSrgj/0i7+ITqIuqftJi1q0s6VvrPxeobTD2IayKum9cElEal8NKtiwGPPpdJMsy0V2OE6ln2pZsloVnJ+fd0qIXYIo0t0mxlpiLRrywtmW7u/WGZq6QFnpX5knXgkPR+MPHGMcOo4ZZRmx7zJjTEvb1l6HPWU8njCdTplMx2gUk8mYjY0NabLeNDz84CGL+ZzlwtP+lBg063MDRbHAtIY6KnhsW1H33N4hTjOuX9tnPLnPyckxT548YbGY+2fEj2kI5a4yRIIxMfQGHg/dhKSjHAJ1XXF8ckQosstHY3Z29tnducbdO69xeTlnsZhzcXnOarkkz6eiL9+0xEnG53/oSxwfHfHw0SO+fvsm88NDsIZzc05Z1SzmlyilyNKUKFLMNqbko1QOaV+sF54BupPKY95WMHAFKuScrOudIqUAJYqbkaZuQsUntNZXRriKyMUsF5f+QAj7TOo3tNJsFiX7q4LdtuXaquDo5k3effCA1WSMUVGXJH2Vh34Vx34VfPJx+HqnqTN4zfAzrr5vaJNpjOHs7KzLswyvkL8ABg2yZWw1RhqmOCextTFgVSdZ3hVe+ntIdISKI9AiygWgnMI2IiI3vveAk9lv0B5ekEYx7O288jnD9akw7grlGwooXN1iI2nQG8q1g9aIdUgC1VgSJ8bWRqAzLS1SjUOrGGtT6lYM0sbmFlpH3aAr3bMDOpoiMsChdVdZfryUZlhgz54/51d/9V/Ie/gJ//JXPsubb77B2dmcf/Ir/5SHDx/TNC17+7vcvnWL5bLqaGlpmvKlL31ZvLjRiMkkJ8tTXrw44Pvf/x7f/Oa3iOOEf+NHf1TghcEi/DivJfSCDRj71UKqcKVpyng8ZmM6YzweM55OieOMslxgTEtVNcznlx43XlLXzaDKsH/vgOGCo20bf2/eUwNP91z5EnyhMSYeQ2zaystJiBpnlKSYVBhNTdtSNw1xnJBlKePxiMlkKpo/WjPKR9RNzenJMaenRxSFHDbOOeG/OznoozgGrQahstcNV6qLEk5PTzg7OyUb5Vy7dp3JdMzrr7/O9vY2x8fHHB4ecnFx4T33bsSvrIp1MayQB1E+ugw1TlfnxFrDZXPJfD7n2bPHbG1tsruzy7Xru1y/cY22bVnO5xwfHbNYLhiNRuzs7vLgjTcoiiXz+YLZbINHDz+iqBo2N7eYz+dUVcXh4SFplpCmCTs7O54BPvRIw9oPhWr9Yyn6exVjagfRSUgqSsLa+aR314xE0fUPDj+XlolAa9iZX/Lg+QtMmuHihHeuX+f45k3Jb6DEAL56562tdTcwjGtMnfAsvBpbl2fpf371wFBKdc3lkyShLEuKongJagF8Hihdgw7l/fw6cBApB8phXNNpvzvnPFGkXwtRFHl52qHTGuZDoOQ2H/Hsy1+kfHzCG+MR7e7Gx4yVXJ8K4+6cwylFZIfuDr2Ub6x9WlAoUoExohREaGxtJNZBk+gJifY9NFGEBrjOicqb9o0dXjXxIQR71c/7y2Ks4ebtm/zQFz/P2dk5OoL7r93jJ37yxzg6OuF/+Ee/wuXlnK2tLTY2Ntje3mY0GnPr1gbj8bjDjIX3Kptf615Q6f7917lx4xabm9vs7uxinedYy5O/BL8Mx1EORIUj6nDpyThiY6aZTsaMRiPG45w4jmiqlvnlJYcvjijrkqJYUHqNdOcr56wdLuaYLMu74qZwNU3Tz6Nzg7/pRaNEylswbuMPVTmshWmhdUyU5aRZwubGFttb26RJmEdDWZYcHb6gbeSArOqq64iDf2argCgm9g3NXYehSHQlWvhNZ7R8bSDOOYrlio8+/JDnzw7Y3tllf3+fBw8e8NZbb3N6csbjxx9xdHxIUawGayMkVv3cOPlZuCXrFATRrIFh7OarYz5J8vLwsOTo6JA4jpmMJKcxzie89tprWGNYrYRHfXZ+Rj4SrrVpDW++/TYoxZOnT3n/w/eIopgkjVDKcvPmTZSSwj1cn1fp56eHj4Q1pRCdo8HacopOg8g/g9YRcRRB6kfCioCAU31hVShSSuKYG8cn7BwdE1c1Lkk4uHuPs81N0V25vByMjSPWWg4PfxBrn7T0WaCOAdNHoVIH0gGsnZ0fPpvMizTGtgSG2NBzVz53sbGxQVmWXF5edt748P2Ct97ndHq6qdYR1jgwhlgLC6qqRHNGR+vYfICEuqQqrziY/Nwp56hwHL19j6Nr11id3eatcV8M+arrU2HcAXRrUcZhImEtKCfwh7agWin/VlrhIo1BeVxTgmClNTqKsU6TMkObmKYpyTLxHoZJFpwjSZLOeIVLKWl7NpzMcK0bejDOMhmP+aU/9ccJyQHnHGcn55ycLPjqV3/EywX3CzxEH9AvfKkk9CFaV7ygyPMRo9GUOE76CXdi2uUm1u8rLIQ4jn0SMGYymXRtypyTLksX52ecnp5SVQV1VaGJui43RV3gXIu1/pz0ycQsS7uFHFgmtcetoyjuFjjdM0APHSm0loPC+fA+YItpmjKbzZjOZmRZTp6NmM02MdZwdn7Kqlwxn59zfnZGWdWUZUVbF+SZhxa6Q96RpbnQALVgusKZH/K75f5kY/fiUwyMroynsKVWT59ycHDA06dP2NvdJ8tHfP7zn6eqH/DU/26xuPQHxWBC/LMPv3fhHnCvWFdXDZB8X9c1bd0yny9I44yNjQ12d3fZ29vj9u3bvtip4PD4iKODA68Gqbhx4zo/9EM/xMOHH1GWJfv7u9y6dat//z69NPjM3hOWQLl/wRDGGD7WGk3QL0itlTS4iJMOM1Yokrrh9rNn7D1/wWGec3Btn8VsRrO7zyhOGA3WsbXSvapt6m6dBYXKwEgJhAjlMW6tNZFKut9rLeymq0ZS8r7Dgr71uciyjM3NzTUI5lVV6kFeOuzfwLhbn1tLFFmfs1Akiez/4YEZDoRQoRueP4yrsLucr8GQsZ0kMT9x6zofTpccXd/ijfR/uQbZ//NdCojAaABHZFwfXvlFo7xxiBHda4dokKtYjIg1jsRkWJvSGAnTt7a3kXeUwFlrKeUWr7M/zQGUhwiaWhKj4v32+uJ9SK04P7vk1371t3FOSvXffOstNmabYA1bsxm2S7y5wd+JkJVSQZ9dNlMvHhQScWGinQ9pG5IkAqX7+/SLOEQBw8SPJNGkOOji8py6qqjqhtVyQbFaiKHzLI5IK9pWC849GmFtS9ta4jgnihKMbWhNSVUbrHFdpa8YekPUlcV7hNHJV7TPD3gPXSmRRR2NRownE2azGdevX2M2m9C2UnCzXCx4+uic07NT5qu5L+P3XHXnSCNFnKXdIT3cTE1TS12CktC2Y/7ghTydXTOk6xFPqHL2xs86nBUDe3Cw5ODwGePRhPPzG1y7do3d3R2mkxnL5ZLzi1POz8+pqqAhIt67JwMN8jm2xxG69Sa4ve6ii2Bq5ZB3yCFbNw3HJ8ecnp3x6NEjtrelc9Lm9gb7u7tsTDeEElnXnJ9f8KUvfZnPfe7zWGM8ZDCw6GvwRNh44fDTnakP/QCcj7SsU1hkH8p0a6IoI4ranuONOF9xnKF1inIR+yfH7B8fsnd8wuMb1/j+9WsUxpAkGeOoNz1hPqRmICJJU1TnyriuC5JzUmMQpIetgbZuMKZa857DewU6axxHnVCgMG4dSrvO8UrTlOlkgjWGi/Nz2qZFDTopDRkzwZm5apBxFmsatJXkM852PSPkdaaDh4fMnXAoDT+ro05q5b121UFqoyjls5sRKEWsr8KD69enwrgr8CFL5LnGEmajHc43RrE2lN8okiQljTLSOGWSj5hkE7JoysHhJVURY5taJsEYDwH4JgZKEUWJN+4h6Rj03SOPsTU9ZsZVvE5Wd9savv3td3DOsbm5yZtvfgalNCpU8bkeNoJ1jFzmcSgmFmRU13HEPgSUKr48z9ne3hZlOger1QpRtRz5ZtzzK7hff3A1TdV5r8HD7vFXCSuNT8ZGUURZVbRtgbU1qIY4SlFEvNrzHEwiEm25FibjnI3ZzLNZhL4YmEh1XfP06VPOzk5ofAVriNastX1HeeV1f5wPTdW6QFh3P17D3FmLacNh0/vPEvr2IfXQ0+rfx2G9cyFG2YtJO8ViMee99+Y8evQROzt7XNu/yd7eHnv7u5RlwdnZKYeHh6xWS29kwth6p8C5tbUwXPcBLblK8dVeMqGPOi1VVXJwIEVhyXPh/U9nm2xubnH79hZJkvDixQu5l+WSpm68QyHvawfjJgdnNKCM9r6sHCw97VgiaOerx2VkxRDH/f12xlgRWcv1kxN2z86YLpf83mff4mh7G2UMuixf4oK/fPWeN6iuh4PYiaiDTrH9GgjrPkRmzokzYo2h9t5/4PQDqEh7J+M6OMdHPtoJuSQdad+rVVQcR6NRJ9QGPRQZDDRKoV0rPQraUOkd04/q+rq76rGHay2v1u9SRNRQGprgRDfrauOeq9enwriDmFqlLUSh6S3EOiFPUrI4Z5ZPmGRjpqMpk2zKOB6T+dJuRcTZRcHhixKtpA+hJPro4k3xTSBNM4KqZD/wiizN2N7e5fj4eGA0+vsbQjv7+/v84T/8hwG6/pP96wBi+iYf3pXr8HCCe+Q/QLwJaYodMxpNPKd2IsY8TjqPuWkbFotlVyATaFifxB6QBW9QKhj34BMNPE0rZU2Bo22tzIHUianO+AGY1kDsMWWvD5KmUowjh08icItvKXh2dsLz5884OTlgtSpo24aqqiVqUX0BRwjxhwnqMKDOKzv2hikknmRTBZXi/ll6497DJrrfREqMV8jpeHvl29uFaR9izPK1risODl5wcnLKZDLh+vVrXL9+g63NB9y//4C6rnj48ENeHB56LrncibEiKdB5aeFeQ9UrDpRIZqguyRlCdNs9q2xxyV80VUNdlcwvFxwfHTObbbC5tcnu7i5bW1udWNh8Ppd6i7LA2uC4yPq3VoxdaBM3zK/4oZV7sQZrw3qV+wvNbcL8WT9vO8sln330SIpvtObbn3mbxWQsb2ktocI15Cv696R7r7BuP/bqoJV+vQ8dqCiWGtY0z7q5dDiskR4CSZJw/fp1tNYsFgtOTk6om7rz1ltrfFNzy3Q67ZKmQ+ciwJlt29C0JbYu0dZgmhbtdaxaLzke2vE5wDrrKaiyf/pnWK+IXXc+fFDpD7Moitagp4+7fiDjrpTaAv4vwA8hK/YvAd8D/jZwH3gI/HvOuTP/+v8E+MuIzf6PnXP/6JM/QDPNZ4yzCeNswna2xWw0Y5yOyOOcNE5lQbgeZw3GQFlwTnF4cEhZVtS1GLw8F31uhpVjSoT/1zHSYLi174m5fAkHHb4uvNf29vaaJ+BfIItuoJS39klhQXqtivF4Qj4WmGJra0sqFb0iY9M0rFYryrKmqZsucSNQ0cuL/1UJ4v5DhZkR8gxBG2ZY/BQ8+7C4It2/zg6EnOJEe175lPF4QpaN2NnZkd6mSnFycsTR0REH5gXnZ+cUxUIEqAahqGzs/nANc2mDYXa91/gqznD3dW0AfE7ADVzQbvCDHIN/T5+U6w4TvJyCP3z7bkHrnxsirbZtuLw85/LynGfPnnHjxk12d3bY2Nzki1/8Ene8rsvz5085Pz8D2xdG9R6664qArAtaNxZjvb/W6WrI1y68dxDTlxI7B01dc3JywunZKc+ePWNnZ4dr164xnc24cfMmSimWyyVPnjzh5OSE1nv04bnEo/R5LB+1+p449E2te+3/YRTinOuirt1VweePTkm15mR3l0e3b9PErMFBgj0HaIPuoFmb15ei5cE0+znqjCbqpdcM+7t2fhQQJwnb29vMZjOyLOPFixcsFguSJFkTMAvjned5x10PubhOhdPfi3IW1bZoK3pISjkponRGKnajiLoStpmKYqI4xnr5ivA5alCsN4Rp+p7GEIrtBBZVa6/7uOsH9dz/M+CXnXO/pJRKgTHwfwT+J+fc31BK/TXgrwF/VSn1eeDPAF9AGmT/j0qpt90n9FHdHG3yiz/0h6XziNJE1jflcL5k3Niu1N1a0QDRaJH5VFBUJccXwu9tmwacX7CIJxeqX+MoIokjCfWDx+Qv41X0mrYh9Gp1A2ytp4UNIYGw8KHzQNYwevncLMsYjcdsbmwwm4knsLGxQRzHFGXB2dkZL1484/LyktWqlLZuGq//kZPEKV0iSa3tFfksbxS7jTfwfMPJbwb3rmxYwP3m6iGivhQaBVmWk2UjtrZ22NnZleKiOEIpYbCsVgXvv/8uq2LlI4oSY1rSNCcKMJAfngBT9OPlS9Cd7Z+p89x9ZtevAZw3ON0zBZ6w18rxuELfQ3d4SZ3zEGqT7kjWVy4L1VQigFBm40D11MZhCTmDXMx8fsF8fsHDJGFnZ49bt26xtbXDG6+/zuuv3+f8/IyPHn7Ekffmw6EWjKt1ojzpfB5J+TPPeswrdJUMwp/OOYzygwFel1z+yFpHURQ8efKEp0+forXuHIednR3u3L7D3u4ubdNydHTEfLGkbUTsTPZAYMn09R4QhRn0B0BPIdZeQXX78pJbZ+dkStPkGY9u3eJ0Z5thdAjiceaZDUi/Z1B5+Ktbsr0nH/bd+lruX+ufupur/oX968M1Go24d+8ek8mEg4MXPHv2ZKAJI1pWYY4DlBi8c3mNj/wGDeVt24I1ouhI4tvwKVIvkdGJ8cWxz0ckaM8y6w/Lfl+EBPHas3tGXahhUd0e1QPplFdfv69xV0ptAD8D/EV/UzVQK6X+BPBz/mX/BfBPgL8K/Angv3TOVcCHSqn3gB8D/sXHfUakI0Zxjra9UZGNp1AegHVGdNKCbnto0YaGg+NTGrHXmNYwGo1Yw50lqyadxrXqGgh0Awg9vc8FzHU9aXeV5yq/X/d2ZbGLHPBkMiHLMrJM2A5DrK4oK56/eJfFYk5RLDs2wDADj2+8pjxPNghXDUM38PCC66MZd8W4h2ccMlnCQSCNLPwGxPkipgmj0YTZdMMXvwirxTlFXdecHB9zOb9gtRKBKef6sREP3PlO9nRjGVQpUf3C7aIvpXD++e1wUyrZ/AF/loOt96y6f4fv/Vmr1XqBm3NBKM5vSNvDY85ZgWK8gZd593p84Q0HnzVcD32YLr9vmpqDg+ccHR0wmWxy7do++9euk49GvP3WZ7l5/TbPnz/l9OyUyheTKS1dmnzcIp9DGCZ/kgWjZYOzoL2xVYNntL7Ct3/ugDGfnkrS9/Gjx0ynEzZ9b9979+7RtpayKlnM58znkrcJWjqyUhQOSVALBDf0smUNXV8uefDsBWWS8PTebex0k3o8ps/xyOu6Q91DOIHz3UVT2u9upzxs82rDPvxZTycc5q0kEgtTFYgH+/v7NE3Du+++y+XlJdD2vQpciFBgPB53DKH1DlL+YDUW20hDIY2Tvrf+4AproQ17EY3Dor020lXDHceaXrSv/3nvlcueaFuRM7XOSuvQtkVFn8yUgR/Mc38DOAL+r0qpLwO/BfzvgevOuef+oZ8rpa75198G/uXg75/4n33sFaZruIkC3qeUlOo6rbFY8XQUned9OV/x7MWRtAxbrtCRJh/lsGYoZdNnWdYbGtd/cEhMhmy2f6bu/q5OMIROMxK6zWYz//cxuRfOSpJEGBXn53zwwQddBWUw8v3h0IdikmQRHn3slQ3D4g12TELDAW1qcE+uw9TlCkY3HDqykC3We7FxljGbiajVzs42SRKRJDnGOFbLgrJc8eLFCz766COWy8KXsFufGJXPjQasAplL8TQkmrBdUiiEyn3k4zHEjikwWBBODnAXiZcmmuuOOE59p6y+zdv6HMkoyFq6usqGujmeg46YSEJkoETOYs3t95601l7bnXVDs46PyuExn5+xWF7w7PlTrl+/ztbmLji4eesmO7s7VFXJ6ekpp2dntKbqvMFgUqW+RXBVY/Hee8DhPevq6i1a03n9yttNgeJAknkNZ6clZ6cnpEnK1uYmd1+7x727t4miiMViyfxyQVlWnJweM59f0vhuZNZIslmS0j6qBa4fH/PWux9wOJvx4Y2brDa3ZY5s70n3EJuMU9saf1grGEBzvnN4N1dDQ/4qmEZd2d/DNRT+NsuyDoI5PT1lsVh4KFWewHjD7BxkWcJ43LNhnHMdxVT2Z+hrIJpIphW1WU3Yu9GV+5QcjlLa9wdYj0aCwxVyIH11eThEpFm6MYEZCBrlabzRWhHUx10/iHGPgR8G/iPn3G8opf4zBIL5uOtVn+heepFSfwX4KwC729syUf6V/eLsE2A+VpWsvAIiTVnXfPD4CcuqYVXWrFYFOzs7HaQSJjnoMadpKh6/c2vrSrxGS7EsfX/IPtwZGvPIZ+q11ty9e9cnqOQeoyji6EgaE0tVW0nrG0FcxenC16AwGLA04cPijWJgSwwSXf5M6uGN/ufBCDEwOuH+jRGDkecj4iRhNB57Lv2IPMuZTCZUZeFhoSMuLxecnpxRViu/oVUXtiZJjHPaj9GVxWx76EOFZKXWAnMN9q40QwnGZz0vEQyH7H3vhQF129C0jjwfdQ2dX2XgP+4K87SWNAS016cPpl5QDh/whxNVOd+BSa+P++A+hmMektdVueTRRw95kR4wnc7Y3t4W6YTJlOl0xvUbNzk+Oebs5Iy6Kv14GELBkDECOsTKeQhtMD5u3bB1UIWPOpWHvEJlcHd4WUdTV5wcH3F2eiQCcHt77O7ucePmdTZmMxxwcnLChw8/4uTkFAdd3iV85mi14rPvvs+kqPhXb3+Wy+mURPU0zqt9SYdYdpjnl8xCd3i+XA169XrZCesjisBFn81m1HXN4eHhK+pXAlwkhUvinK1HlqFzmDyzER0pf4i6VpgxUqnbe9/DBt7K68RIScXLB5bUiiTeYXt5HctBpNFRRBRLFb/Wiey18LCfcP0gxv0J8MQ59xv++/8aMe4HSqmb3mu/CRwOXn938Pd3gGdX39Q59zeBvwlw/95dJ8kwCVEDNffqJgo4PCga6/jgyTPOVxVt61gtV53hDRoQglnJxhyPR37PeqzdrdOtnHWdAY89xp1lmRf56rW767qmqkpOTo5YLucsl8tOZyV0ezLGrTfYRnUHlgruFSE5Il6hHACRwCSuD3/79wh3Ks/QYc/hdYODCgTKSNK0q4rL0ozNjRlKQWskwbqYX3J8cEhZFtR15eEqjTGOumm9NkgfIkZRiKx8E41BpOVc688W8cgi7TXQTeuby/SFPPKe3nMbRGqOQbGPRRqv4Aj64daGTkaux5w9TTI8ewcRsQ7PhPHsYBm8qqGTw9SFtRG8fz9N1in6zkVhP/XRiLMichYO3H786ayxtB0UmYPpdMruzi67uzvMJhOu7V2jrmoePX7Mcrng/OJsoFHiHeArsJq11sNpYU/0nztcJ5GmhzU9HKIIjoWM3XI5Z7G65PGjD0mzMXt7e2xubjOZTLl18wYbGxsi2FYsKcsCipIbh4dsXlxQpgkffO5zLHd2SK84FFcP/t7jXv/+6tUfWP17DHWehpHS+iXrZH9/j1u3blJVFc+fP+8owuv7Cb+fY8bjyVov2n5N2+6rMQZnat/yUZzMxOsVDVk0sr56p1KiWuv7I9AZ/QDvSIOUepBM1YOxkfeLk75xuNDBZRC1+p8Bc3fOvVBKPVZKfcY59z3gF4Df8//9BeBv+K9/z//J3wf+llLqP0USqm8B3/h9P8dXZGmPQ66dSn5XicaKomoN7z16zIvTc4qmZb6Yo7RiNputGRytNZHWTCYjkjTuPAIZ6KgTUtrc3GRjY4OmaRiNpPVZURQURcFysaQ1jUi6ehU8a1vKsmC1Wqx5JFGs0JFQ3yLfXEQm3hHHPkTTwrUPvSf7ylT/evh9wq1+bDo4yye2BOPPiaOkk3GVw6jCGsPBwSHL5SVFsaIqSxR4Te3gvSoCIylNh+XZL3tSw2hENo/2hkP7BLYkqULDb/EgZdGGBGLUtR3zob4acr0DFupxdy2c4c4LIsL4qMfiiK4Y+EAp7M6+7tb9wek8BIMcANZaVBRU/cIw+8NC9aMdDIMjQE3K0wT738mwBCtmu9oN62B+cc7i8oKDF8/Y29vj+vUbjPIJ9++9RpolnF2ccXBwwMHBgdc4ch67DRDcYB68dygGRXXr7eW8EH7sfGvK0MBEBSE+jXWWsljx+PEjnjx5SpIkTCYzdnZ2mM5mbGVbKLXNzjvvkDQNMfC9Bw94MR6Hmxp4vuFQvrJu/firT1jf/UHl8ARp1lldrzbsSRKxsbHD7u4u8/klz58/GzSo1mvjopSw5iaTyVokFmAYMdjihFhroS3Rrm9oHSnfzHsASQ77BMRXMPaA3ffRI94+BW+/7eBXrSMcjta0PsKVSE5ZUU41LtzXJxt2+MHZMv8R8P/wTJkPgH8f2SV/Ryn1l4FHwJ/yD/odpdTfQYx/C/yH7hOYMq+6homy7nvk1FyWJd/76CMOzy9ZFjXLokI7xe7OTteQt9e2gMl00pXip2na8dLFAGYdDn54eChqh00DznWG3A0W6RBbDQnYkAyNokgStpHuIJVwqofXWus8jh6ebH3hytUXG70a4RKPI89zEt/lZzIVKeEsyzCtoaxq6qbm4vJC1ByLJWWxoq4KCf29vx9r4dsqHTxovxDVlbP1yoYKm0/GW4yEdYbQzT7YwuAlXoVCgl/deadKKgmN6z2TIT7eb4zgIalubvzLu3l3SPPzYVXqOj7eqRQN7kPuyRrz0pCHv4+09mtB7s0GjRIHTvXe27CnqHLDSGEQkuMoyoLHTx7z5OkTZrMtbt64wd1797hz5zY3b97k9OSUjx494uj4SDjzXcQpm7wfkyE81RsQ5Q/KQHHEY8MddNadD35MnRJGmve667pmtTrk7OyUza0tRqMxb0YRd8uSKo753s//PLMsw3raZxDbC0ZUdQlpf4BcWVO/39VFUPRf+7lbT0zmec7GxhSc5tGjR5TlauDVrk+o1tKPNgibdfg2DLTrNc61tE2DbQrP85f30n6tQhB/G46369ADwDO4DNqJ46J89bQOCXP/bJ1Ok8fgm6aRNWytd0Bk/londSWB7y59DD7++oGMu3PuW8CPvuJXv/Axr//rwF//Qd57eHVelwtVnt6AIBH8clXwzvvvc3h5yWJZsVgsidOUze1dkXXtEoi646Jfv7HP9vY24/GYqqpQSlFVFcWq5ODgULxaa71x6u6Eq2Glf65u8QdPeche0ZHynrn8eZi0YBylYu3lCVkPZ+WPeyZOL1SUpilZljMZT0jTjDiOSLOUphGe88mJVHwO5XgBr4EyZC/IM4oHOuCF4xC2iPKwxzqnuW0Dw0CBkmSdeNQhAhCvPFQShyph5znkfbhrvTJjP65YK9V9vrjFWhnHzpNmqDyoAO8RBzhCScCnPS/ZWeX1Z+za/PUemfJdvTwrx6neqR94eCG6kda+fuNbR6S9aJ11A286UDQ9C4Jw3944KZ/I9hFCELE6vzhmvjjj+cFTbt++w/7+DZIk487tu1zbv05VV5ydngnTpqx88VEvoCfLfiilMSQmBPriep1AsJ1D2NAFob7uwLM411JXBVsnx4yeH3Acxxx88Yu0UQRty/3797u6jFAQJBFtiAxkra0b61c7DOs/83ka5YDWj3Ew7iJPMJvNuHbtGmW5YrFcsFgsvLPVR37DK1AcBQ5p18TvoDfWzhpwLc5UmKbtoh8pHtLYQfQcrtZ3Veq46kqjrOHw2WPKsubmvddQcZgL0zl+Mn+hUE2kVZRShP5kzmfUAzLlnHTJUmja38dl/nRUqPrTswvXvOeulIS+BsXp2TkfPHzCxapkuahYzJdEOmJzutGV5SpUh+dub297BkjCYrHg9FS01IMgkDW9R3UVl103gutXZ8AUnRfehXK+qCLSV/DP7tR42RNfCxcRGlgcxWS5UJ2k633UGfi6qr2uuu9qjxVtlStUzfXP6IdV/uHx6MFml/vQOGL/++AJ+s0y8KIlMYfHxIOx6KEJ65N9cj71DJlhOG6t65pYX7lbAr/c+uSiUqJPXjdt915DiE2oi6ozzMPkdQgCnfP+jz+Eokh1hkw2U99ZaO2wDZFNd97Ls4bo0tFHDt3neC9VnBVFa4ynaFp0pES7RPfefBDHOjsT2mKavsf29i43b95kY2ODmZpw4/oNVqsVz58/5+jokOViwdCDDc87xHwF48WzwHrnRPuuZqjBQUCg7wWIwa9IFZGXFT/9rd9mPpnyu5/7HMdti378mNj3JZAk8YStra2OJVZXNWUlstvW9RFGf6SvO1DDtdpj9o4ANw0j2TiK2N7e4bXXXsPhOPtQiAz92n/ZWw9NZ8IVhMk6ZwW//5zF2IoYQQpaK+yeIEfQVUuHA9H1QnS9UqTsm2ZV8OLxE85O5FB+8wtf6ooqZQ/0tSfgexerkIQdQJ7hYAtIn3NE+pO9dvi0GHf60Ntazx8FlHa0LuLpwTFPnh+yKmqWxYrVqiJOEra2tnwFmZde1aIOOZ1NSLOYxVKaHvTGaz1sFc+ilwsWzep+Yw89uKEhDCFuFPXsEABFRC8lysDIhGeE4YIO9Mc0TUUlcbqB1qLBohTM53NhN7QNTdNSliV1U/UTLu/KK23k2hVC2RAy+4WDV93UGqe07wSjsQaRLZWt7qlckTdaqntGyebL5gu9VNvWYkxFGkegQ+QiLoa1DChj630l1xZ879KLAfIHQZ5l/pnFgwsecbf9BwdIz2AQPDkkE31mwOOdPVtHKWEjdPorXkhMkshB2tc/q59FKYBy4uU5h1ISmSklhXJCjUQOTV+MglVYZwZNnn3CVoXnN1TVihcvCg4Pn7Kzs8ONG7fYmO2QJCn3XrvD9Rv7XF4sODo64vLyHMkNtAODLnPctq5bH4FGGTxi3a0FX1/hCwVDklkpKbp5+/iUzx684Gy6wfP9fY43N2WsraWx1neQOu80WLJ05HM/I0ajiWgH1RVNU1GWJaZdj4hlSNepxj0Tq4ctJeJyTCYTtrevkWU5z5494fjkuGsbCENYyB8Eccx0Ou2ahHcNZq4kQnGiSktdEiMiYNYKpJLGMToSLNzF4gQI+aBHF8Jebxppbh1pTTaa8qUvfQltSp4dXmJsQCTWRc4EarE4J03Z1eBZOh0dJDowoUfw4O8/7vrUGHflQbkh42G5rHj0/JDT8zmroma+WGJsSz7KmU6nV6AYaZ03nU67fpJDL3ZorIc/CxBI7y14U+F6j34dNpFriCP33mTAGvufB+8xZM/DogoNM/b39zusz1k4PT2jaaAoliyW0oPU+mrbcC+qt2adkf+kib6aTApGvsddnU/69QyUbuNdeT7r+hoEMfC+AIlWCsRaT3GM1hXwwqEYGlrjPchuxMOmRAysYNW+IYTnddtQ6OM9ZMkVeCfUKglfXxW5MCgQ8+yftgljaDqPX9bg+nwP366D14JOiPxwsI5k/r1vD8i6tM5z/lX/s6swxPraClKycHh4yMnJKVk6YWdnh53dbQ/RJdy/f5+yWLFcLjg9PeFyfonz3PYh3BLec239uJedllBD6ACtImZ1w2uHh+Rlybcf3OHhrVvoKyXywzleLBYs3ALtIcvJeEyaZWzMZuhok7quKYuKoig6OPTVUfJgLfsDN89Tbt26wc7OLhfnS549fcZytegi51fNu7SHHPl8V0hm9glmoIP2XFNhmwrlexnoSBHFKViJfKyQ1qWymUA11ehYd4MmTUk0kY6ZjDfJdEZ1aSnmK+7cvkthnW/c4rr8oDQjl2cObLtwQFtrO7jnalT2vxrPvfeaQFkwTvHk+IjHT5+zKmoWi0LwSeTknvoEItBpLSRpzGw2W2u2cXXDhJ8PL9eFjBbJzkvJ9fC1Afddb2LgIR3XG4Y1L957BXmes7u7y+7ubuc9rFYrFosFs9kMrTXPnz9ntVpRVRVNUxKKm1AKZ71HLXdC74H3RvilkuW153NeEGxYbbeGU8i4q6gregnP1OUhLBgM1tiOXdMnCgMsI1z61jQEkaRwP8Mht9ZKRapWGLRogIeo2ypCcbpSA4Pk0wIW4/Wtwxj00hI+sPWKfH3iVanIV756Zg6aKErlPjvMUuMZnH7uDYH+iLWDRGN4pqAF4NGgTivFCh+9cxQkMnS2x1i1Xqcu+mAJNxQ2u7JOrbVU5ZIXzwtOjg/Y2Nxkd2efJI6Ik4jXH7zB/vXrXJxf8OzpU5bLhe+MFdZ8D9kIbKAJMgtDL9f59ZBEKTqK2ZmfYmLNbz54nafX91DO4AydYRquuX5tSfRdFA1VtfL9R3M2NrbIsxGjfMxsNqOqqo5i2bZN10krzEewmFrH3Lx+nd3dHdIs5vDwiIMXPskctCZY36tBdiFN0w7yEsPbRzYi2tbgLLSNQdkShZU6F6zkTXzjkciJY2aNVIl3DmUs8FaSJKBAO+MhmYZ6WaDTjMloj3E6gXhCbGsWq1OM6zXq5RC33boY3mvv2Qt0p7QiVq9Wk3zV9akw7i64iVozv7zk/SfPOLw4Z1kULC4XWOsYjydsbW51BnIYWqVZxsbGdI1HuvbeV752XlxwYdYwd7X2uvDv4YbsXx+YAJoo0iRJTp6P2N3dlYYKZdmpJI7HY1arFcfHx75tW8uzZ8+8gNAQjw4hacDaAg4JhISnP+n7zP7L1XyvjloGh4QKXq8VSMYE6CTgnG49evH/Wwtl6b8PiVStYxR98VWQzo20JooST92LuttwvgF4l3zzHnmQRFVKeSUAR2ManMOHxEo0bnR/72HTDg/grjhEWRwGnPfi1g55X8rvHNaIquD6ePWXc8NOU/2q6USeCBGPjGHbrqsf9tBIz9LxTvYazrq+xuVWrDXUteXo8JCT41Om0xlb21tY54i0dMm6e+8eG7MpFxcXPHr0kMViueZ0WOvQynbL+KrXK5TemMCfT5uWJkqwoVLWBUeov9eeVHB1rCx1XVFVJfP5OUmSMhrNxMj7wqHxOCfLZU+fn19wfn5O09QoNEkSc+3aTa5du0bTNHz3u++wWhU4G/I667CeRMdRp+QYnku8YxnLnkUjyvmYlkQHo+0EqvTS18rnYYxtMe069Dd0IpsgIOgL/tJYYYszSpMQpTlKTWlaoVpGSkMkh4G11tMv+8NsaGvC2Ib7j6PEP/cQFv3461Nh3AFWdcvTF0c8e3HE5XLB5WJBUa4AR+5P+yH1MAzweDwmH+UE7ZCr11WjDldPvFAw1G/MVxlI2YTR+mePJkymEzY3N3HWSYs4382lrmsuLi58154F7777Lm1be36qLIJhVVofiobPDt8Py/uH8Ip4293vnBPP1fnKRJ8IDcnYkFyTwyLABuGZtR/XGLy+hyQp+wbi2slmUJ23ZMXjd944OQdaDaQJwPlQU/tIoG3r7n4DLVT0fmJ6mp/w1qUITJJLoawf38w7SSK64hIPCwWsani4yzOHsNezJ3S/FuTgMUQqJghYSQQyPOCH4++4yhMdGpdubrRnPAUtoMihIu81u9afmz46IUhcB/ndqDNC6weGWrsPa1vm83Pm83MOXjxle3uPvb198ixjd3eXu3fv8eCNN3lx8JyHD9/n9PRUDJufR4m8XAdZyQgpYq3RymAdvPHiOV/+6CO+e/smTnmP1ke2EgVEDCU7ZPyHyeywlkM5f0Fdt1xw7huxT3HOkY8yRqMxeT5md2eXs7NzyrJiPB7jHLz/wXsih9C1c5R9cjUiTHweLiRJh9F62AMiuaywrcHVFdBgTUOQR+jkj9vWH9oyl1YpWmuxpiVPUonGoriDApVYa79mNUQRDYbSSCXrqqlYXhwxyiGajJBmPK1fp8rPqUR/IRIMB3+I3F0ja6Kp27W18XHXp8K4143hd955l8WioChLzi7OqZuGKNLMNmaM8nH3wNIkQJPnIyaTiT9h1yvD4GXPdfj1Zbyz90x7r5XO3ioUcRJ33sbGxgZbW1tMxrMOz1utVrRtzaNHj16iJPbermzM4SEz9NbWJ6unjwl/fP2Aunp1RUI+edZ5fIhed6g2FcfFP3/3VnJvMgYedjEtgluE+zNYa7yUQeyjlXWZVGctdJtdYWyIBnoPQ6uBd+qL1SItDAgRfhs264i6g0AOH1+05NkIckD22H7vQb6MaYfoIqwDgZLW14fyzx6MQWewVWBGQFgYYT57SltfAWoH0UMURUKz7KAsfy+eq9w60c1J4l7T5Oo8O+uuFB6FdSFz1zSGg4MDjo9PmM1mGGOYjKc+uad48OABd+/e7Rp+N2W1tie6wMLJWpeHtVy7vODaxQX3j495sr+Dcz0WHMchYd17m6/aX8No14U1gmK1WlEUK5RSlGXBaJSRphnX9m90cKVSUt1bleXAm+0/c/g5URR1bJ3hnAbpkaGmkwKMqYj9nFpENjng3dYK/Oi3g9y/6w+QcEgEqATvKGgPQSocaZZha3lNay2tacjHKQklrWkpm1YiLs9V796vi8R9/G7dS+s5wDTDPhKvuj4Vxr2qaw6Pz1iulrStDEaa52xvbq5ltQO7ZDqdMJ6MJHTErnWYCdfL2Pr6Rr/6M/F+xQCiNHHS9yE1rWFre4vxaIyx0vHn+bMDjH2KdMipWC5XtO1638VXwTpKxYPvhzrjgbHTe+bODTH0wHu/yt+1ncFART3icuXw6j8jLEBFgJWca33ir/GYumDmygkFTLjsoisjWhfaeziuK+YJBtyZ3siF00OEjkwAvFBAHDaCk0ihNaabT60jD9WEsTO0TYuOeEkSIWwE0fiIuj6pw1LyXqZ2YNi7MEYOw76HnBj4Yds2Y9dL0/tT0ecIvCcukIfyFYzWE2CCdyysHXqYWIIdPy/yvIph0VO3FgYJxmH01c+pXE3TcnJScX5+zmQy4ebNm2xtbeGcYjrdYDLZYH//BpcXFxwfH7NYXhD0fZxVqMjnHuTT+e17r/H1d9/jJ975Pt988y2KNMJagd6ci7u5CeO0diANjO9QPTXAZv1+1V6RUubx+PiE0WjCdDLrnIednT3AUZYVq2JFWa5egiSm06k0tPGNba4mi3sWkaJta+pmiXHW06KFmppl0ouVqOe8d9x3P6eRlsND1ofH4bUUN7VtS2sMWZQSxQmJ1lirMEqRRAmoEXmiuawrmtYQeV2awH4JUJIc8taLhvU67kOWTNfb9hOuT4Vxb9uWs4tzrLWkacrejjBI1GCTJolgioKnJQT6VMAGw3XVqMsPr3x7xdDHccwoHzOdzrDW+k5IY6Ioom1bqqpiMZeON03T0DSNtLtiHQZ6VTIT1iste+9PPLIgpDPsSiSvXfcog+EZJjL7z/FQin+sHoIJ7033nqFCMXiASgUa3qB7p3JoFRHFkuzLssxXw3mqY9tiamlh5nSgmPr30DEa1X2GHnDhAaF6qUHpv5/DUHXprMV2E+a8t27RUYh0hoUfPZwmibP1BFtvQCxx3Hs53aGLp0a+Miry0YdP5A/hkW79uPCyEJVJCN21thweLp6K0n+G671ErXynKb12z91rB7cV4IY+OlMdVBnWV+hNsFwuybKMvb099vf3OxhzOp2yf+0aR0cHnJ+fs1jMJcr0UYoMq+b9W7f4e1/7Gj/13e/ypYcf8Y3PfGawFvv7CeOxHqX2Y9of0iFiXW9ePtyPIvtRcnZ2KpXYWSYS1LMpSZKS5xnGSkK2KIqOTTIajbr3GDJMwj30zTFEMjdWXpjNBvhMMPHgQMpr+0ilalqiOJbqUe8ABWlkfNTWttKFqWkaLoqWjw4OuXHtFuM0I9LgiFGqwlpDlued/tLQIQyfL0yZhpDnCuM7rML//a5PhXEPDzebzTqoBeh6GOpIsbEhsrq+YNyHdz0cM7xeDg/FS+1QbK1J04SNTZG7nUwmpKm07CvLUtrZzeecnV9gTEtdSXcke+Vz5DNe/txXPd8r4RTVP/vHMV6c64skOq2LwevEIPTGJfxNjxHjkZWhpxuy76FDFQMjJE2845ju52W58vfn/CGiUFiiJEEMoEFrn1TUsbd3fpMrhzONGDoF6NBuIPTd9N4z2jdkCd6zFhaJG3RtcobIM2yck4NIeSlWHRBz6+GXDtpx3guSe4hULB5q2+K6KKan5PX0vFCoJrK7YUyH2K21QnFsbWDPW7SKwUU++dZrhqxfrruvcAgLtS3GmNqHXnHnCGincOqKI8F6In193UgUaK2lWK149OhDnj9/yvb2Lvv7+12V5sbmDpPJJlVVcXF5yfnFCXVdIlx9B7Hm+fY2yywjsqKpgr7qhLx88L3sbPn9p4ID0ss8r2+Z9XXcNDXGtCxXS87OpbWhNIWP/R5OOwdsiK8P+5sOvV1J8Ja05YrYVzi3rSHSCh1FnS0Jh0NYD9ZaklhUW4e7uBepAxVp0ZpRsGoN5arm4ZNH/PPf/F12tjbZ3t4iiTL+jc/uSfSOBRV1MGW47+GBbZ0l1hFxFCiXApcRIMJXm5vu+lQYd6U1+/v73Yk5xE/H4zGzjemabng4TcP33RWg8isLPo5jIp1ItZ+nYcWJTFYogDg/P+9wwBD2SQg6NLrr970eoveft/77l/89vOGr0M3HvXZ4YL3q2YMheAlr7tgi6qXwVP7OijFSQ68xcLV7rG/YpQciwQsRaCmOBw2Wleq98y4k979zjjZ0tukaP8hKddZKxavqqWqhAjjw3NVgfFX3v8GzDn5mnb1iK5z3sAzORQTEvYc4hlzw8Eb9Dro6rkOnIiJASv5QGzgfVw/d3rNVBKjNeQkDa1v6Hqf9ffcNOgYMG3oq41WDKmvW+qS6fFbTNBwfH3NycsJ4PObmzZts7+zhImk/t7e3x97eHhcX5yxXSxYLUVN8ePMmv/n222ytVuzN55xsbXXP8XK0sz5O678fUmKH+3P9++FYDcfZWsv5ec3l5UXX/i5UnAbINsAXAb8OPPHAH7fW0DZCJhjSDXUUd9FFHxmtEzSSOPZEhT46iXR/INguwvOFlFPNL/zoA977/kOen8/ZGs9YFZec11N0HGMHh1C4//Dv8BlZmuJTS1iLwEcI9VLB7+u9fyqMe5iQMLhRJE2XZ7MZeS5MGId7acD7Ba26kFcP8KnRaMxolHcYXJqlrJZLyqqimddrDTT6hSghcWfknIhhDcP94bUOnfQ/+7jE59VruOHXcV3WNtDVDTz8/dXPH/5++O++e3vfSs45jfbVqcPPU8rL3LpA9wyfL5i0FFw6zztXqM5b9QeNR0gUdB2WrA3MAHBK9SbRGnotm/UwP0ghy+FhezgnjJemgxO0kkpL62zHinF2naansGgtfHTpeenvWQXYwK8BJd2NjFUoLZLDHRYe9GqunMECLzkf/YQoxxEMdKh2FadBvFkwHSwmhU4951k+L5KxcWEc12mt62sA6Dz6oDg6yMn4+14sLnn33Uvy/AmbW5vs7+1L0tE69vb22Wg2qcqSy8sLirLgcH+f2+++y2ZZcqrWHZ6Pi0qvQp9XHZ2r0M3V1199n/DVGOM7gEkh0GQyYWNjY22PhN910ZXr4V2FOHsmVEg7OnZZpDXGrR+aIYkakrLhPbv/jKwTNKRZ4pP9sv5tteLutU3u37nB5mzMSVVThjEb2LveYVMdVBTw9yjSaBVjrKFtDUkaYZwVuQzzya77p8K4hyvoM0ynU8bjcffgvZe1bshCMiKJpdfo1tYWZVl2p3Wo/Ly4uMS6tsPPXwXlDK/+dx2o+rGv/zic/ZPe/1Wf96q/+7gNMlzIV0PioWc/fI9w0ncL3T/fQL5r8Dw9/isFFgywczHQTq17V8ZvKIVgqiGZGPj6wWgOowOlPSTl51jo531hVr8pvXftO/iEVnhB80URxMr6g8liRRjsyhz9f9s71xjLsuuu/9Y+596qruqe7nm2Z+xo7CjGkhOkJJgoAYQizCMJkSdfiIyICMIofEDi9YHYsoTEh0gOoAghJJCVgBLIAyskgCIh8oCIL3kQAgQHZ4iNjT0P9/RMd1d3dXf1vefsxYe11t773HuruycZT1W17hr11L3nnsd+nbXX47/WirQA9oOperZRNDht35TCEZikc6m6hq7TjH8Ze4Xc2L9bGKVBdY2FW7qL1QhDh7BStTDVmNfWfBHtXDH3aIKUQ39AFcYBy4kiYeYKnLedc3R0l3tXjrhx/Trnz5/n8jue4/z584zjyM7OLpffcY5hWLKTM2/cvInM5xP/0erYtsdXf9vE4Dd9Xj3/QYy+jT7d6Bfx8+28wWokjwPaZI1U1fIWiII0kOucc4E2h4RtfMdNJ6OVzlRRsmuAUXLv2vUDehK7sx1u3zlkoYJ2niFWqwbQwruj3cMwoNl8WJ0ODLk6YNtArPvRqWDuAiV1wPnz51cckFU6NxNAR9/P2Dt3jp3dHYZhyWw2Y7kYOLp3l5wzy2HB7TsLhuWySd0LLQOraQKCmbVQrni++L/N0smmxbt50Yco2zLPzZL9VHVvFNaARKX1RR/XARsZe1HjUyKpAp07kVpThfXbzAOVARpjTdXdqh6chDEyzZ52ACXNDEc9HqnZxiWy+QVzarMEuq4ZsyIVbhhjGX4GHQeSGL7dChbMUB2IjI6K8TYQJM1IeL3ZhG0EzlytBZ5HZXR7NyCqCNUEFEEyZQBCvEiG2Q9zh8gUzhoSdp20QOr0jKOZXboSjhu/Y4nm1GZJiLXpL7qOIF3sNkV7Sp6crjjYRYvmFLBhdYdLMINo5+Aai7gJaxyXXL9+jZuHt9jf2+fJJ59ib38PkRl9N+P5rLzv5Ze583V/kEt7+7WQfLNeV+k4przKxDYdP05DXb0OKNWSQqpuE4FBNesE+smK6SwZc7YsraNrf0VrNGGBWLOavBhLvYeqlmNZMyPKfDb3GVXGnMnDkqs3bpLoebKfcefoELn4JMmBCV2SIr33nsk0Z19PnaHV+vkMzcpieY9uPmM2n5HH4YHCadCpYO6pS8WbD3UxRFrdQNG0FVP298+D4FWEFpZUy3Hla8FMAtq+hPYUjOE2KZyKquiXrUDoJlLaMcx1s4pq6na9twWCbFqsrfRRLtD2+/oz15/XHq+Y7Nbeq26SCUFS0aZiTOSMNv3S+Ennqr5zr8A8M2LFpZVup2d0x3NXbJghLSYqIigTUayWZqF2dV1SxduYIaRQdRNGbMyA0oFjjVPXkzqT8C1upwa8aFRWQhCxpHGqioWyV+m2tqXauquJhbIRTTJQ5mAMFfYXLFvDdyA9UQRksmZUi+ZR5z92u8BmDx5UE5DW2EkiEZrDWLVqtlONz4AE4YSsY239ynnJvcUdXn7lJSti89gl9vb20es32L91yNMvv8KND/xhdnfPWSnJo7vcu3eXB63DtTUd47WifR63URynDbfnxrxFFbZwpIazNTRAsA2hwyshjUOBWuQhCqgruIQsCJpq3pdi+vE4ja7rrCC2+wRTjHeacem59zAMwjDeYjg8ZO/CE6iGo1ecAwUEtpbbVFUD0QXYokvMumRCSGO1OBM2976fFQ9+qFAXLlgVGFVzfuzu7nLjxg2uXr2KqnLz5s2COXex5760qk7ijsFQc1cXXnvd/RbV/aSKMOtsWK+lLZtsj1Pbt/o7Hp75bu38tZdighaByHopmkBNi1FVsy8WTaIWEagvPC6djO7MVAvIQUDN0y/BxPGMgmJwx5QSUhAfNdWDiGVaNKnXzSnxrJJ7PeakmnBSinw14eylkZLVcsAgVlsu7CokusAs+/gZIjE0GSEipuJljeyWrYO51uONPDZjuVdoO0nEN4gwvUBFMVUmN11/fnazgSBhCjM8eRcRkrmOIVSJtLzfZfPS0g7TcmwnsrQCcwMmiFU0s83G72UrxO3HAmRuHNzg8PCQ1+8dceXSJV551zsBZWdnzs7OnAuPnQc1MMLh7cNJbMCqBnmcyWT1+PpvdfymJGUcWmoje6Md1aFqJsIksfElxr4jdR3iwt/oeYCGAlrILAezrYdZRDF0jfq6au37rYloORjSbnd/l8Nuh3PdDEUtRbdHu6cuot7VN6Ja8COrkrqOXXfmxjqLNXksY3E6Fcx9Pp/z/PPPW1m727fZ37cMeMvlkldeeaVUeamZ5NpJbcW+es/NzHZdOt9Ex11r121GCGy6rkr+q/DFzc9sKSL0WmmxhiSvt28qJbmaX/R0gGTSdxpM6hYQLCPiqjMqKr2E6m+SKLQaTgA6CiywydjYd53lwVgaVDAgjX5FM47QFlKukigYs2vzlogXSZiabsLWHUmkqsHDkDyq9R1QtW0ha5hdLM1q+B3iyRPmpK4hiG0KOZumkfzF9FPM2akOzQw7rtS2tG0LabuOg1KTZfn+hCCYI9mKlqzbV8M0pLGRaHX2Rmpmu4/5KiQgfz4oEd3buamvE/FNSkqCNs0jl658mS/PZ1y5fUi+dcD+/r7HPuxw9849Hn/8Cfb3LZXAwc3rVmdgk9ARK2CDwFT7s4n513ww9lNlW5O0AlBs8GFGiU2xZFYcE8vFEsYR1YGlo1w6wfLup4jmtrnNg9n1R80sh5GU7L6zWY8mWzlR8W5V6HviqScQlHt3j+guPMOYOvpkxYdQy8I55qGxoVP6YH4dhz0KxW1ldniDXI4PqNbxQOYuIu8D/nVz6KuBvwv8mB9/N/AF4LtV9bpf8zHgIz4rf11V/+P9npFz5vXXLS9z3/ccHBzwxhtvsFgsJtkM123aGtZQbLzW1cBVqbo9Z9Xk0l67YRzWjq0uwnWGC8GMS/m5lXseZ2ec2HILs4i6ouuby+Qe0sQ0+oc8DoBJ7WajHOg7CnY+pEBjDFOopSQpRaBN8p7mkxZfdBMp2xmNmV1Gx5/HeFtwVKu1lGmUdk4ca1wPU1FTkX4g0EymwQ3DwvLRUFPr5gyaU93o1BhnOFfreNaAG9QcnJY7xDSMseDSo1RirtWLYg/1lzbUipwDUUTTr5gvgI5xsILXRYMUf5slu5O2XeN+r4nq4miyMkh+b82OShqZdbOJRGuJ2arTehxHZ+pd2fBBeOzuHZ48OABVbogwXr/OzZs3OX/+fAn37/vebdkj/ewphmHg7p273D26y3JZ8wnVPqxDKVc12KnNPuzoIehMN8e24tlqvEgwe3O+uk2+m9GlrpQvRKSsK82mOWX1TUiwGr19T++bRHXkDpSaEL6AW3x8SplhMZAk0bcm52aTTY2GKIgHQ43MZgESMFRXRo7dFI+jhymQ/SLw9T5YHfAy8LPAR4FfUtVPiMhH/fv3i8j7gQ8DX4sVyP5FEfkDep86qsMwlEoqxzU+1J6mXfFpMtnHUSymdoCOY5Kri+tBDH9to3BVuJgsxOqqBjZ5YkE5pi0whZnVYanPavOftJ/bbHm4tJlVGXVhpiw3reRxpEt9qdIegUqR3CyuF7FAociFvm6WCrOFm1KwUonDaAy3hSrWfkdv/Dsgjhs2c4xFEMYlebI2QktwLchLCKpL+8NgvhRNLp1nRT0TpfS2eSlq5dQAJLkCYdJv6sRz9fhzwl+BpT8Yx4E8VtSM2fqTazj1/BIMlMGsHesCRM4Dw6jsdHMs8Vtm1IEkHeKQNxqNRhXymMA3A0VKAirc5m7+DXd65wFE6LrdIhFWhImWvEOWWsLGIuy+4zhyZ2+PLzz7LMuu81TGZvq4ceMGBwcH7OzscOniJWaznSKQ7O+dZz7b4Wu+5r3cvHWDK1eucPv27Qnk+H6mmbpG4p2Kfq2bOCcV3Px7hdBaX46OjiomPiXoepJm+i57VDPMZ7NiAcoKKpmjxb3im5g3PKH6WaJCWhPUpnWt5GFAdSTryOxcVzOdJjONjcNyRciB5ClEhmGwjbNxEMe9c84lzcL96M2aZT4IfE5V/5+IvAB8qx//UeCXge8HXgB+SlXvAZ8Xkc8C3wT8yvG3rVWYbNCmIcyrEvumhbG2KShTCZbjF1G5ZMPCaq9bPXedyQW0UAyGl+Kl9Oc1JpDVe7bqY9vnNlHUans2jdeme4YEo+oSjqvunVjlqNb00UrecR1U51xd2Kney8O3yxxRkQeRzTHsRJKySyv2W3Kmn7PlXQnpv6IehAjFtqAmRxk0DjKL5EuGKEoxNuZMtQ0nHIvm/LXslsGIcUk3ImpNEksdpm3k0A0jF3tsuG2On6lpqQobvuGlMPtUlFYwguhvuxqjf5aKwU0E6kkZimYVY2xOwCqxRsRzR2T3DG2rXSMVoVHXSEjBcV7XdagIX3zuOd7x+uu849o1vvTMM6WfgEnod4+Yzebs7Z3jiSceZ7m0+JHXXnuNJ598gq/72sscHh7y8ssvc3DzRslOucqpWwFpImRpHPPx8TnVnFkuB4ZhbBBMMmG01bG6JKU5hiQy53REnOdhAGe8Bp0ZSdS0voG4i/uuMvnV4zGHtq4Vy920rpW0+WGqGVYC8cuYI122aVNJKvLpQel+4c0z9w8DP+mfL6vqqz4Zr4rIM378ncCvNte85MeOJ11VV6fMfKqi1bzM9qIfh59NjaEKWlRG0MMw+eOYcZw3+V2rGSV27xrNU6Wv1f5U5qtri2RTW1bVz/Y+7YvabgJtemEJFS8bvju899GOungj5em0zeZUqrA/SUJeLkkaUYKCaE/fz1kuIwOhJ9ZCSB0ugfhcE9j57FBNiM1MfW1kV5d3dswEY05L34yibWWD0XJ9FCEpQgC2IQVDtyNu23RMfs4uXXUOE81iqKAG4aLl+kBHeFFkwL2yjUZTmb0xnbGuEUmkFH4kN1d5q7q+Q0csMEqVcQwJlrohEZkpUzFVtcwndcYspvNa104wiXWTZ6VbFy6wt1jw+sWLK79XiXocl9y6teDo6Dbz+Q7z+S59P+P27TssFkv6vueZZ57hwoULHBwccHh4aMiWZLOy6f2q76e42StsW/HsjsW9Jfv7WrSC2Ww2MdOERrlYHLktvSeljmHh6Y/dcXlvsWQ2n5PEA6BcA7N5qfEN7YaxOn6t7d+AAgYWyXlkljrP0mlQSWne61nXs8xD8cxJMpOh72sMnrguiZlHj7MkrNJDM3cRmQMfAj72oFM3HFvjoCLyfcD3geWUWc+Zcv8JDylzkyQvIiWApW1WeTHLd/tbHqNF4F95ZnV2rbetXi/uYLH/5Uk7axvXpWxT7xuTjjKRkmFdU1i1V66bk6JdVXLuut5/6UBHy1eitti6sjAFS9YV/bXCwN4LL5MHeLIrQ+CoFc1QC6Lp+xmiieUypJrIz65l3EVcWvHhMiYvBNyympYSyEiXMVtoFlIX0qlXkSpIKAqDD00p67Lex5Rlk2qX2Uum4S/MCLHZqEld9l9vZh+pG82qpqNZijbixiGENggpGHvtf9WOKH1pUxIb8kRIJeVymPcilbI5kiU2g2YjsbXjAS9dZPZMa+sFKk48mGh7TnweRXjl8mUAuokA4nOZEgGZtajvgdu377JcDDz99FPlPru7u6hqYfS3bt3i5q2bHB3d8TbU92FivgnUlHo/kmlUgJsZYTabuxaTS8BRMOHZbMbdu7ddiLFx6/oZaIZxhHHk3rAke+Hwzm3t0Y5A2qxqsmH2qRu+me0CfilEbhq1PE6dFZfppKZKiBwyuJZeuERsxCK2GbiWmdWQO2aff+uCmL4d+E1VveLfr4jIsy61Pwu85sdfAr6que5dwCurN1PVTwKfBLh8+bJrx/fHuYa9qZzjk58894adtxlVEGH0a9I2FGYaTHnahMBFK8Wr58dxya2w0wJb3Gy336QlVOnapMxqvqmh+i21KQpWJfXVe4pUiQKg6+aoGrKl65Pj2v0eXdfsakrqkgdVJDpP6pVSQlOFaNpfg+otx8AFq1WbUYM8WuGL2DSSpwvIhvSPDUVHxiFMO9H+Dks1bBA90Rmw8Gf2pBT49MRyzMxnPVL8BRREgToDb7UoVTXz0OA4Y8zBSqpQOjQyBro0rNJsImEi8GAiSSUHiJZ1YmNk6859AoV5CYEAqTDI7LBGoVPLPUKGUa2IiqQIN/d3we/bCcW8kFJoAonZrJqwJPUESinWS5ufqAgRGtqY+Ia9vnZD6Ijv2VMnVIRVLZRx6/CAw9s3SSlx8bFLPPPM5ZIaZHd3lwsXLnDpziWuXr1quPm7dycxBtIw2NgQTSNWaupNOLq7YD6fIUmLPbrdzGaznsPDJctlz2y2g+bEKHO6NNjazop0PYqn0nU+EpHu1oZ4R91ksmLbTw44qA5+yDqSXOIehgUqiaPFPWazedGIQ6Mq5SK1Ji4rYoKnK6hABRot9Xh6M8z9z1NNMgD/Hvhe4BP+9981x39CRH4Ic6i+F/j1B918FfVSTRCVSa1L6Lg9U2rVG0mspsXdZOKRELFDB27u23XNghKvlCP2ItvS8p3aca7q5ggabaJVxaMPqwzfn+i79qq5xSRCbcwO1g+T3nCno3gf2sr2RZUNSUyrLXI1Zeh0nOxv/Y2yqCRNVU+R+Ny7DdXTQSTD0yesXJtJq2NpGxMzRTBzizzOxOZl/anz0LsmMZS5zt7P1GWYVOFqkEOuueRsLHzqgPJ14OOUc3Y79Oh57eOlbrNSTtcJhEMv/BYaU1eRK0VrizVdhZRoQ4nE1WoCaE0oIf378nJtaCSYaln3ZcNJZR1D8jkUpuvf/rbQQY3xaIWWiUAT62TaF1Q98rZ916owNY4j165d58aNA/b29rh48SKXLl1ib2+P/f19dnd3uXbtWpG6Dw4OPAK2zmnXmR5ThB+tms5isWA2mxXfz3w+m9QNjrVayjO6L0GHkSS2Trvc4D2EYrqMuQpmbE7Ryo9q1GogZ0xTMC3clkzfWxbIvuuKQAVTjH4rsLXvaWyY+EYXDt52rI+jh2LuIrIH/CngrzaHPwF8SkQ+AnwR+HP+wN8WkU8B/xt7G/+a3gcpU8mDSghbYu1Ay+BbqUGxDdx2eACXbtaw05O+2N9UuXpghM2RETu+lJc5db6Ys7iUFG1LjgsOyb/i0IOh5sBJF5NBnvbB5zdevliw1l9xc030hdI2CakRiooa7WqZepgThGR45mzRjIGwiDEtzzYdqmEORqvpU0uUZ6qO3zAFSO4Q6RxBASqCeD5zVTVzkAhZhd4z8qGjC2NhFxWXHHvQwVTp3Dvjz5Y3pYyjFikqbM+R+AwsaKWNNvYRIvuc2vsZxYmDUYV9PFaaQRKzmwnifq2jTNWLYXv9zeobirlJZc5S6hG13CF0NjdgNty25ufUFo7XjB2g8wRjVP+H+UhqFHDOStfbfIemtfoe1M/1nSuClS++8s64E1l83eAMjPIOmMCjaFm3rSyTc+bWrVscHh5y9erVwuSXSytyc+HChfL8iDiP30KwSR4zEcKdMdihCHop9UT6hfg3jgPz+Y7d894Re3sXICWW48hyea/MVfh2bP6rYzalvrTBhIDK6CuyyMY7bPAGkrJnM2bXCEyLGMdWSJpGK0emyEDL2Pr14LLspjixwiALP/c4eijmrqp3gCdXjr2BoWc2nf8DwA88zL2DJqYSjSCT+8McyyIjpKGA+U3vu+lZVf4OqToVNTQ2FEMhxCKq164mKqK8vHFvKRJ0u0nFs1qb5eQe7Uf/fZO5pTynGbNV1EM8CwIxEVCyisqp4y6gib6rWoM0jCLOq9/F4itCeyrM3sdJQTrD9irAwvK54Awjq9vnG1t5aCddJ/T9zG6SMHx6OLLG3sfIMvjlbE5YYZgw74BTxmZvL31uXuBqkgjeXcctIkSrhjOFuImbrBQzoyWSmM9iuQxnWzVTTNddSMidV/TxPP39rDhNI9ldSJ1TlEtoQQmJXEXJ6r+GCWM2mxXmHhtDqyWsvg/Te0/jJVqzX1nzyOT3oGKzT53VaJUmBsAFNh3DV5JYLBZcvXqV69evc/HiRZ5++mlyzly/fh3V6hjd3d1luVx6AONYhaKJlmvS++65HcKP0WqoIh3nzp0zzWC5ZLa0fEiznR0W6llS3ZmtY9VKNo2TpQif1YhX/y0qk5V8MV0Po5lVxjEz6MBsd17a1M5vfA+JPUw+IbAEo4++l9iffH+buzxItH87SERuAS+edDveYnoKeP2kG/EW0rY/p58etT5t+/Ngel5Vn970w6lIPwC8qKofOOlGvJUkIr/xKPVp25/TT49an7b9+f1RevApW9rSlra0pbNGW+a+pS1taUuPIJ0W5v7Jk27AV4AetT5t+3P66VHr07Y/vw86FQ7VLW1pS1va0ltLp0Vy39KWtrSlLb2FdOLMXUS+TUReFJHPiqUOPvUkIl8lIv9ZRD4jIr8tIn/Djz8hIr8gIr/rfx9vrvmY9/FFEfkzJ9f640lEOhH57yLyc/79rPfnkoj8tIj8js/Vt5zlPonI3/L19mkR+UkR2T1L/RGRfy4ir4nIp5tjb7r9IvKHROR/+W//WDYFs7xNdEyf/oGvud8SkZ8VkUvNb29fnyJo4ST+YXXuPocVAJkD/xN4/0m26SHb/Szwjf75AvB/gPcDfx/4qB//KPCD/vn93rcd4D3e5+6k+7GhX38b+Ang5/z7We/PjwJ/xT/PgUtntU9YZtXPA+f8+6eAv3SW+gP8ceAbgU83x950+7F0Jt+ChUj9B+DbT1mf/jTQ++cfPKk+nbTk/k3AZ1X1/6rqAvgpLB/8qSZVfVVVf9M/3wI+g718L2AMBf/7Xf75BTzHvap+Hogc96eGRORdwJ8Ffrg5fJb78xj24v0IgKouVPUGZ7hPWFzKORHpgT0sId+Z6Y+q/hfg2srhN9V+sSSFj6nqr6hxxR9rrnnbaVOfVPXnVTUSrv8qljwR3uY+nTRzfyfwpeb7g3O/nzISkXcD3wD8Gis57oE2x/1p7+c/Av4O07SaZ7k/Xw1cBf6Fm5p+WET2OaN9UtWXgX+I5XF6FThQ1Z/njPanoTfb/nf659Xjp5X+MiaJw9vcp5Nm7pvsSmcGviMi54F/A/xNVb15v1M3HDs1/RSR7wReU9X/9rCXbDh2avrj1GPq8j9V1W8AbmNq/3F0qvvktugXMHX+OWBfRL7nfpdsOHZq+vMQdFz7z0y/ROTjWPLEH49DG077ivXppJn7Q+V+P40kIjOMsf+4qv6MH77iKhbye8hxf4L0R4EPicgXMNPYnxCRf8XZ7Q9YG19S1V/z7z+NMfuz2qc/CXxeVa+q6hL4GeCPcHb7E/Rm2/8S1czRHj9VJCLfC3wn8Bfc1AJvc59Omrn/V+C9IvIesUpPH8bywZ9qck/2jwCfUdUfan6KHPewnuP+wyKyIyLv4SFz3L9dpKofU9V3qeq7sTn4T6r6PZzR/gCo6peBL4nI+/zQB7E01Ge1T18EvllE9nz9fRDz9ZzV/gS9qfa76eaWiHyzj8NfbK45FSQi34bVk/6QWkbdoLe3TyflZW48y9+BoU0+B3z8pNvzkG3+Y5ja9FvA//B/34GlRf4l4Hf97xPNNR/3Pr7ICXr3H6Jv30pFy5zp/gBfD/yGz9O/BR4/y30C/h7wO8CngX+JoS7OTH+wYj+vAktMWv3I76X9wAd8DD4H/BM8GPMU9emzmG09eMM/O4k+bSNUt7SlLW3pEaSTNstsaUtb2tKWvgK0Ze5b2tKWtvQI0pa5b2lLW9rSI0hb5r6lLW1pS48gbZn7lra0pS09grRl7lva0pa29AjSlrlvaUtb2tIjSFvmvqUtbWlLjyD9f7vXBW+2Sg5eAAAAAElFTkSuQmCC", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# draw overlay\n", + "%matplotlib inline\n", + "import matplotlib.pyplot as plt\n", + "\n", + "plt.figure()\n", + "\n", + "traffic_world.overlay_trajectory(cam_id, trajectory)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "b1f6fc81", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(121, 4)\n", + "(121, 4)\n", + "(113, 4)\n", + "incorrect length: 113\n", + "(58, 4)\n", + "incorrect length: 58\n", + "output video files ./output/trafficScene_cam1car-1-a29b573e-70a0-4b6c-b19d-4f453867ce4f.mp4,./output/trafficScene_cam1car-4-a29b573e-70a0-4b6c-b19d-4f453867ce4f.mp4,./output/trafficScene_cam1car-14-a29b573e-70a0-4b6c-b19d-4f453867ce4f.mp4,./output/trafficScene_cam1car-21-a29b573e-70a0-4b6c-b19d-4f453867ce4f.mp4\n" + ] + } + ], + "source": [ + "# render tracking video\n", + "filtered_world.get_video([cam_id])" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "f8ae640c", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Times are: [(datetime.datetime(2021, 6, 8, 7, 10, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 59, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 1, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 2, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 3, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 6, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 7, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 8, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 9, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 10, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 14, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 15, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 27, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 28, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 59, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 1, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 2, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 3, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 6, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 7, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 8, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 9, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 10, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 14, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 15, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 27, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 59, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 1, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 2, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 3, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 6, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 7, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 8, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 9, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 10, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 14, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 15, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 27, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 28, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 59, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 1, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 2, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 3, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 6, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 7, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 8, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 9, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 10, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 14, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 15, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 27, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 28, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 59, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 1, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 2, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 3, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 6, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 7, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 8, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 9, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 14, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 59, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 1, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 2, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 3, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 9, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 10, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 59, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 1, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 2, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 3, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 6, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 7, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 8, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 9, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 10, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 14, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 15, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 27, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 28, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 6, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 14, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 15, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 27, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 28, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 59, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 1, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 14, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 15, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 27, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 28, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 59, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 1, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 2, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 3, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 6, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 7, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 8, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 9, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 10, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 13, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 14, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 15, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 27, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 28, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 58, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 10, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 11, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 12, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 6, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 7, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 8, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 10, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 4, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 5, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 15, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 16, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 17, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 18, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 19, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 20, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 21, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 22, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 23, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 24, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 25, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 26, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 27, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 28, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 33, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 29, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 30, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 35, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 36, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 37, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 38, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 39, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 40, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 41, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 42, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 43, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 44, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 45, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 46, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 47, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 48, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 49, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 50, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 51, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 52, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 53, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 54, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 55, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 56, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 57, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 31, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 32, tzinfo=datetime.timezone.utc),), (datetime.datetime(2021, 6, 8, 7, 11, 34, tzinfo=datetime.timezone.utc),)]\n" + ] + } + ], + "source": [ + "times = filtered_world.get_time()\n", + "print(\"Times are:\", times)" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "d2973e30", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Bbox geos are: [(2.0351596988934304, 2.400814869380193, 1.0, 2.3896888485676846, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.3896888485676846, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.3896888485676846, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0351596988934304, 2.400814869380193, 1.0, 2.3896888485676846, 2.7472250308939685, 1.0), (2.0351596988934304, 2.3960036171369463, 1.0, 2.3896888485676846, 2.7520362831372154, 1.0), (2.032453369506604, 2.3960036171369463, 1.0, 2.3896888485676846, 2.7472250308939685, 1.0), (2.0351596988934304, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7472250308939685, 1.0), (2.0351596988934304, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7472250308939685, 1.0), (2.0351596988934304, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7472250308939685, 1.0), (2.0351596988934304, 2.3911923648936995, 1.0, 2.3869825191808585, 2.7472250308939685, 1.0), (2.032453369506604, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0351596988934304, 2.3911923648936995, 1.0, 2.3815698604072058, 2.7472250308939685, 1.0), (2.0351596988934304, 2.3911923648936995, 1.0, 2.3815698604072058, 2.7472250308939685, 1.0), (2.0351596988934304, 2.3911923648936995, 1.0, 2.3815698604072058, 2.7472250308939685, 1.0), (2.032453369506604, 2.3911923648936995, 1.0, 2.384276189794032, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.3869825191808585, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.384276189794032, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.032453369506604, 2.3960036171369463, 1.0, 2.3896888485676846, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.032453369506604, 2.3960036171369463, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.400814869380193, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.400814869380193, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.0297470401197772, 2.400814869380193, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3896888485676846, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3896888485676846, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.384276189794032, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.3896888485676846, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.397807836728164, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3960036171369463, 1.0, 2.392395177954511, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.397807836728164, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.3951015073413373, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3951015073413373, 2.7424137786507217, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.397807836728164, 2.7424137786507217, 1.0), (2.0297470401197772, 2.376758608163959, 1.0, 2.397807836728164, 2.7424137786507217, 1.0), (2.0297470401197772, 2.376758608163959, 1.0, 2.392395177954511, 2.7424137786507217, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.392395177954511, 2.7424137786507217, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.392395177954511, 2.7424137786507217, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.392395177954511, 2.7424137786507217, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.392395177954511, 2.7424137786507217, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.3896888485676846, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3911923648936995, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.376758608163959, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.376758608163959, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.376758608163959, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3869825191808585, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3896888485676846, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3896888485676846, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.392395177954511, 2.7424137786507217, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.392395177954511, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3951015073413373, 2.7472250308939685, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3951015073413373, 2.7520362831372154, 1.0), (2.0297470401197772, 2.3863811126504526, 1.0, 2.3951015073413373, 2.7472250308939685, 1.0), (3.1934686764551166, 2.978165138569819, 1.0, 3.269245899286255, 3.3245753000835943, 1.0), (3.19076234706829, 2.9829763908130658, 1.0, 3.2611269111257757, 3.329386552326841, 1.0), (3.182643358907811, 2.9925988952995595, 1.0, 3.2584205817389496, 3.339009056813335, 1.0), (3.182643358907811, 2.9925988952995595, 1.0, 3.253007922965297, 3.339009056813335, 1.0), (3.1772307001341584, 3.002221399786053, 1.0, 3.247595264191644, 3.3486315612998285, 1.0), (3.169111711973679, 3.002221399786053, 1.0, 3.2448889348048175, 3.339009056813335, 1.0), (3.1609927238132, 3.002221399786053, 1.0, 3.242182605417991, 3.3534428135430754, 1.0), (3.155580065039547, 3.002221399786053, 1.0, 3.2367699466443383, 3.3630653180295695, 1.0), (3.152873735652721, 3.0070326520293, 1.0, 3.2286509584838594, 3.3678765702728164, 1.0), (3.1501674062658944, 3.016655156515794, 1.0, 3.225944629097033, 3.3630653180295695, 1.0), (3.1474610768790683, 3.021466408759041, 1.0, 3.217825640936554, 3.3582540657863227, 1.0), (3.1420484181054156, 3.026277661002288, 1.0, 3.2124129821629013, 3.3726878225160632, 1.0), (3.1339294299449363, 3.0310889132455348, 1.0, 3.2097066527760747, 3.3967440837322975, 1.0), (3.1339294299449363, 3.0310889132455348, 1.0, 3.2097066527760747, 3.415989092705285, 1.0), (3.1339294299449363, 3.0407114177320285, 1.0, 3.204293994002422, 3.4063665882187912, 1.0), (3.1285167711712836, 3.050333922218522, 1.0, 3.1988813352287693, 3.415989092705285, 1.0), (3.1203977830108047, 3.050333922218522, 1.0, 3.1961750058419427, 3.415989092705285, 1.0), (3.1149851242371516, 3.059956426705016, 1.0, 3.19076234706829, 3.4256115971917787, 1.0), (3.1122787948503254, 3.055145174461769, 1.0, 3.188056017681464, 3.4304228494350255, 1.0), (3.1068661360766727, 3.055145174461769, 1.0, 3.1772307001341584, 3.4304228494350255, 1.0), (3.104159806689846, 3.059956426705016, 1.0, 3.174524370747332, 3.4256115971917787, 1.0), (3.10145347730302, 3.0647676789482627, 1.0, 3.1718180413605057, 3.4352341016782724, 1.0), (3.0960408185293673, 3.0695789311915096, 1.0, 3.1718180413605057, 3.4352341016782724, 1.0), (3.0960408185293673, 3.0695789311915096, 1.0, 3.166405382586853, 3.4448566061647665, 1.0), (3.087921830368888, 3.0792014356780033, 1.0, 3.169111711973679, 3.4544791106512602, 1.0), (3.085215500982062, 3.088823940164497, 1.0, 3.166405382586853, 3.4544791106512602, 1.0), (3.085215500982062, 3.0984464446509907, 1.0, 3.166405382586853, 3.4544791106512602, 1.0), (3.085215500982062, 3.103257696894238, 1.0, 3.1609927238132, 3.4544791106512602, 1.0), (3.079802842208409, 3.108068949137485, 1.0, 3.1609927238132, 3.459290362894507, 1.0), (3.0770965128215826, 3.1128802013807317, 1.0, 3.1582863944263737, 3.459290362894507, 1.0), (3.0743901834347565, 3.1225027058672254, 1.0, 3.155580065039547, 3.459290362894507, 1.0), (3.07168385404793, 3.132125210353719, 1.0, 3.152873735652721, 3.459290362894507, 1.0), (3.07168385404793, 3.136936462596966, 1.0, 3.152873735652721, 3.459290362894507, 1.0), (3.07168385404793, 3.136936462596966, 1.0, 3.152873735652721, 3.459290362894507, 1.0), (3.07168385404793, 3.1465589670834597, 1.0, 3.152873735652721, 3.459290362894507, 1.0), (3.07168385404793, 3.1609927238132003, 1.0, 3.152873735652721, 3.459290362894507, 1.0), (3.0743901834347565, 3.165803976056447, 1.0, 3.155580065039547, 3.459290362894507, 1.0), (3.0743901834347565, 3.175426480542941, 1.0, 3.155580065039547, 3.459290362894507, 1.0), (3.07168385404793, 3.1850489850294346, 1.0, 3.1582863944263737, 3.459290362894507, 1.0), (3.0743901834347565, 3.1994827417591756, 1.0, 3.1609927238132, 3.459290362894507, 1.0), (3.0770965128215826, 3.2042939940024224, 1.0, 3.1582863944263737, 3.4544791106512602, 1.0), (3.079802842208409, 3.213916498488916, 1.0, 3.166405382586853, 3.459290362894507, 1.0), (3.0825091715952353, 3.218727750732163, 1.0, 3.169111711973679, 3.459290362894507, 1.0), (3.087921830368888, 3.2283502552186567, 1.0, 3.174524370747332, 3.459290362894507, 1.0), (3.0933344891425407, 3.2427840119483973, 1.0, 3.1799370295209846, 3.459290362894507, 1.0), (3.0960408185293673, 3.252406516434891, 1.0, 3.182643358907811, 3.459290362894507, 1.0), (3.10145347730302, 3.252406516434891, 1.0, 3.1934686764551166, 3.459290362894507, 1.0), (3.1068661360766727, 3.271651525407879, 1.0, 3.1988813352287693, 3.4544791106512602, 1.0), (3.1149851242371516, 3.2908965343808663, 1.0, 3.2070003233892486, 3.4496678584080134, 1.0), (3.1203977830108047, 3.30051903886736, 1.0, 3.2124129821629013, 3.4448566061647665, 1.0), (3.139342088718589, 3.3101415433538537, 1.0, 3.2313572878706855, 3.4544791106512602, 1.0), (3.1420484181054156, 3.3197640478403474, 1.0, 3.2448889348048175, 3.4544791106512602, 1.0), (3.1582863944263737, 3.329386552326841, 1.0, 3.255714252352123, 3.4544791106512602, 1.0), (3.169111711973679, 3.329386552326841, 1.0, 3.2665395698994284, 3.4544791106512602, 1.0), (3.19076234706829, 3.3438203090565817, 1.0, 3.277364887446734, 3.4544791106512602, 1.0), (3.19076234706829, 3.3438203090565817, 1.0, 3.277364887446734, 3.4544791106512602, 1.0), (3.2367699466443383, 3.37749907475931, 1.0, 3.290896534380866, 3.4544791106512602, 1.0), (3.2367699466443383, 3.37749907475931, 1.0, 3.290896534380866, 3.4544791106512602, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.4384027775305595, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.5732794835417299, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.5732794835417299, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3038493579199046, 1.0, 2.435696448143733, 1.5732794835417299, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (2.3707445428599, 1.3086606101631515, 1.0, 2.435696448143733, 1.568468231298483, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7604785576982099, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.45466333698683015, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7604785576982099, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3575135991909715, 1.0, 0.7577722283113836, 2.780903796596697, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.45466333698683015, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3478910947044778, 1.0, 0.7577722283113836, 2.7905263010831907, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3527023469477246, 1.0, 0.7577722283113836, 2.785715048839944, 1.0), (0.4519570076000038, 2.3478910947044778, 1.0, 0.7577722283113836, 2.7905263010831907, 1.0), (0.4519570076000038, 2.3478910947044778, 1.0, 0.7550658989245572, 2.7905263010831907, 1.0), (0.4519570076000038, 2.3478910947044778, 1.0, 0.7577722283113836, 2.7905263010831907, 1.0), (0.4519570076000038, 2.3478910947044778, 1.0, 0.7577722283113836, 2.7905263010831907, 1.0), (0.4519570076000038, 2.3478910947044778, 1.0, 0.7577722283113836, 2.7905263010831907, 1.0), (0.4519570076000038, 2.3478910947044778, 1.0, 0.7577722283113836, 2.7905263010831907, 1.0), (0.549384865525753, 1.924500897298752, 1.0, 0.6603443703856342, 2.0399709511366773, 1.0), (0.549384865525753, 1.924500897298752, 1.0, 0.6603443703856342, 2.0399709511366773, 1.0), (0.549384865525753, 1.9196896450555052, 1.0, 0.6576380409988078, 2.0351596988934304, 1.0), (0.5439722067521003, 1.9148783928122584, 1.0, 0.6576380409988078, 2.0399709511366773, 1.0), (0.541265877365274, 1.9148783928122584, 1.0, 0.6576380409988078, 2.0399709511366773, 1.0), (0.5385595479784476, 1.9148783928122584, 1.0, 0.6549317116119815, 2.0399709511366773, 1.0), (0.5250279010443157, 1.9196896450555052, 1.0, 0.6549317116119815, 2.0351596988934304, 1.0), (0.519615242270663, 1.9196896450555052, 1.0, 0.6549317116119815, 2.0351596988934304, 1.0), (0.519615242270663, 1.9148783928122584, 1.0, 0.6522253822251551, 2.0351596988934304, 1.0), (0.5169089128838367, 1.9196896450555052, 1.0, 0.6495190528383288, 2.0351596988934304, 1.0), (0.5250279010443157, 1.924500897298752, 1.0, 0.6468127234515024, 2.0351596988934304, 1.0), (0.519615242270663, 1.924500897298752, 1.0, 0.6441063940646761, 2.0399709511366773, 1.0), (0.5250279010443157, 1.929312149541999, 1.0, 0.6386937352910234, 2.0351596988934304, 1.0), (0.511496254110184, 1.9196896450555052, 1.0, 0.6278684177437178, 2.0351596988934304, 1.0), (0.49796460717605207, 1.924500897298752, 1.0, 0.6224557589700651, 2.0303484466501835, 1.0), (0.49255194840239935, 1.929312149541999, 1.0, 0.6170431001964123, 2.0351596988934304, 1.0), (0.48984561901557294, 1.929312149541999, 1.0, 0.6143367708095859, 2.0351596988934304, 1.0), (0.4763139720814411, 1.929312149541999, 1.0, 0.6116304414227596, 2.0351596988934304, 1.0), (0.47360764269461475, 1.934123401785246, 1.0, 0.6062177826491069, 2.0303484466501835, 1.0), (0.4654886545341356, 1.929312149541999, 1.0, 0.6008051238754542, 2.0303484466501835, 1.0), (0.4654886545341356, 1.929312149541999, 1.0, 0.5953924651018014, 2.0303484466501835, 1.0), (0.47902030146826746, 1.929312149541999, 1.0, 0.5953924651018014, 2.0303484466501835, 1.0), (0.446544348826351, 1.929312149541999, 1.0, 0.571035500620364, 2.0351596988934304, 1.0), (0.4384253606658719, 1.924500897298752, 1.0, 0.571035500620364, 2.0351596988934304, 1.0), (0.4276000431185665, 1.924500897298752, 1.0, 0.5737418300071904, 2.0303484466501835, 1.0), (0.4221873843449137, 1.924500897298752, 1.0, 0.5683291712335377, 2.0351596988934304, 1.0), (0.41677472557126094, 1.9196896450555052, 1.0, 0.5629165124598849, 2.0351596988934304, 1.0), (0.40865573741078187, 1.9196896450555052, 1.0, 0.5575038536862322, 2.0351596988934304, 1.0), (0.3924177610898236, 1.9196896450555052, 1.0, 0.5466785361389267, 2.0351596988934304, 1.0), (0.3870051023161709, 1.9196896450555052, 1.0, 0.541265877365274, 2.0351596988934304, 1.0), (0.38159244354251814, 1.9148783928122584, 1.0, 0.5358532185916213, 2.0303484466501835, 1.0), (0.373473455382039, 1.9100671405690115, 1.0, 0.5304405598179684, 2.0303484466501835, 1.0), (0.3626481378347336, 1.9148783928122584, 1.0, 0.519615242270663, 2.0303484466501835, 1.0), (0.38159244354251814, 1.924500897298752, 1.0, 0.5142025834970103, 2.0303484466501835, 1.0), (0.40053674925030275, 1.9196896450555052, 1.0, 0.5142025834970103, 2.0303484466501835, 1.0), (0.3301721851928171, 1.924500897298752, 1.0, 0.5006709365628784, 2.0303484466501835, 1.0), (0.40053674925030275, 1.924500897298752, 1.0, 0.5087899247233576, 2.0255371944069367, 1.0), (0.3328785145796435, 1.924500897298752, 1.0, 0.48443296024192023, 2.0303484466501835, 1.0), (0.35182282028742806, 1.924500897298752, 1.0, 0.4763139720814411, 2.0255371944069367, 1.0), (0.35182282028742806, 1.924500897298752, 1.0, 0.4763139720814411, 2.0255371944069367, 1.0), (0.3193468676455117, 1.924500897298752, 1.0, 0.4384253606658719, 2.02072594216369, 1.0), (0.3328785145796435, 1.924500897298752, 1.0, 0.4276000431185665, 2.0255371944069367, 1.0), (0.3301721851928171, 1.924500897298752, 1.0, 0.41948105495808735, 2.02072594216369, 1.0), (0.3301721851928171, 1.924500897298752, 1.0, 0.41948105495808735, 2.02072594216369, 1.0), (0.29228357377724795, 1.9196896450555052, 1.0, 0.3870051023161709, 2.0399709511366773, 1.0), (0.2895772443904216, 1.924500897298752, 1.0, 0.3788861141556918, 2.0303484466501835, 1.0), (0.2895772443904216, 1.9196896450555052, 1.0, 0.373473455382039, 2.0255371944069367, 1.0), (0.2895772443904216, 1.9196896450555052, 1.0, 0.373473455382039, 2.0255371944069367, 1.0), (0.02976962325509007, 1.929312149541999, 1.0, 0.2598076211353315, 2.054404707866418, 1.0), (0.0270632938682637, 1.929312149541999, 1.0, 0.2354506566538942, 2.049593455623171, 1.0), (0.02165063509461096, 1.934123401785246, 1.0, 0.22462533910658872, 2.054404707866418, 1.0), (0.00541265877365274, 1.9389346540284929, 1.0, 0.21109369217245685, 2.054404707866418, 1.0), (0.00541265877365274, 1.9389346540284929, 1.0, 0.21109369217245685, 2.054404707866418, 1.0), (0.0, 1.9629909152447271, 1.0, 0.10554684608622843, 2.0640272123529115, 1.0), (0.0, 1.967802167487974, 1.0, 0.09201519915209658, 2.0640272123529115, 1.0), (0.0, 1.9629909152447271, 1.0, 0.08389621099161747, 2.0592159601096647, 1.0), (0.0, 1.9629909152447271, 1.0, 0.08389621099161747, 2.0592159601096647, 1.0), (0.02976962325509007, 2.09770597805564, 1.0, 0.07848355221796473, 2.323834833488243, 1.0), (0.02976962325509007, 2.09770597805564, 1.0, 0.07848355221796473, 2.319023581244996, 1.0), (0.02976962325509007, 2.09770597805564, 1.0, 0.07577722283113836, 2.319023581244996, 1.0), (0.02976962325509007, 2.09770597805564, 1.0, 0.07848355221796473, 2.319023581244996, 1.0), (0.0270632938682637, 2.09770597805564, 1.0, 0.0811898816047911, 2.319023581244996, 1.0), (0.0270632938682637, 2.09770597805564, 1.0, 0.0811898816047911, 2.323834833488243, 1.0), (0.02435696448143733, 2.09770597805564, 1.0, 0.0811898816047911, 2.323834833488243, 1.0), (0.02435696448143733, 2.09770597805564, 1.0, 0.0811898816047911, 2.319023581244996, 1.0), (0.0270632938682637, 2.09770597805564, 1.0, 0.0811898816047911, 2.319023581244996, 1.0), (0.0270632938682637, 2.09770597805564, 1.0, 0.0811898816047911, 2.319023581244996, 1.0), (0.0270632938682637, 2.09770597805564, 1.0, 0.0811898816047911, 2.3094010767585025, 1.0), (0.0270632938682637, 2.092894725812393, 1.0, 0.07848355221796473, 2.3094010767585025, 1.0), (0.02976962325509007, 2.09770597805564, 1.0, 0.07848355221796473, 2.3094010767585025, 1.0), (0.02976962325509007, 2.09770597805564, 1.0, 0.07848355221796473, 2.3094010767585025, 1.0), (0.02976962325509007, 2.09770597805564, 1.0, 0.07848355221796473, 2.3045898245152556, 1.0), (0.03247595264191644, 2.102517230298887, 1.0, 0.07848355221796473, 2.3045898245152556, 1.0), (0.03247595264191644, 2.102517230298887, 1.0, 0.0811898816047911, 2.3045898245152556, 1.0), (0.03247595264191644, 2.09770597805564, 1.0, 0.0811898816047911, 2.3045898245152556, 1.0), (0.03247595264191644, 2.09770597805564, 1.0, 0.0811898816047911, 2.2997785722720088, 1.0), (0.02976962325509007, 2.09770597805564, 1.0, 0.0811898816047911, 2.2997785722720088, 1.0), (0.02976962325509007, 2.09770597805564, 1.0, 0.08389621099161747, 2.3045898245152556, 1.0), (0.0270632938682637, 2.092894725812393, 1.0, 0.08389621099161747, 2.3045898245152556, 1.0), (0.0270632938682637, 2.092894725812393, 1.0, 0.08389621099161747, 2.3045898245152556, 1.0), (0.0270632938682637, 2.0880834735691463, 1.0, 0.08389621099161747, 2.2997785722720088, 1.0), (0.02976962325509007, 2.0880834735691463, 1.0, 0.0811898816047911, 2.2997785722720088, 1.0), (0.02976962325509007, 2.092894725812393, 1.0, 0.0811898816047911, 2.294967320028762, 1.0), (0.0270632938682637, 2.092894725812393, 1.0, 0.08389621099161747, 2.294967320028762, 1.0), (0.0270632938682637, 2.0880834735691463, 1.0, 0.08389621099161747, 2.2997785722720088, 1.0), (0.02976962325509007, 2.092894725812393, 1.0, 0.0811898816047911, 2.290156067785515, 1.0), (0.02976962325509007, 2.092894725812393, 1.0, 0.0811898816047911, 2.290156067785515, 1.0), (0.02976962325509007, 2.092894725812393, 1.0, 0.0811898816047911, 2.290156067785515, 1.0), (0.03247595264191644, 2.0880834735691463, 1.0, 0.08389621099161747, 2.2997785722720088, 1.0), (0.03518228202874281, 2.092894725812393, 1.0, 0.08389621099161747, 2.290156067785515, 1.0), (0.03518228202874281, 2.092894725812393, 1.0, 0.0811898816047911, 2.290156067785515, 1.0), (0.03788861141556918, 2.092894725812393, 1.0, 0.0811898816047911, 2.290156067785515, 1.0), (0.03788861141556918, 2.092894725812393, 1.0, 0.0811898816047911, 2.290156067785515, 1.0), (0.03247595264191644, 2.0880834735691463, 1.0, 0.08389621099161747, 2.3094010767585025, 1.0), (0.03518228202874281, 2.0880834735691463, 1.0, 0.08389621099161747, 2.3094010767585025, 1.0), (0.03788861141556918, 2.092894725812393, 1.0, 0.0811898816047911, 2.2997785722720088, 1.0), (0.03518228202874281, 2.092894725812393, 1.0, 0.0811898816047911, 2.3094010767585025, 1.0), (0.03518228202874281, 2.092894725812393, 1.0, 0.0811898816047911, 2.3094010767585025, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4722431864335455, 1.0, 1.21514189468504, 1.6213920059741986, 1.0), (1.1583089775616864, 1.4722431864335455, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4722431864335455, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4674319341902986, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4674319341902986, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4722431864335455, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4674319341902986, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4674319341902986, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4674319341902986, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.1583089775616864, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.21514189468504, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4674319341902986, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (1.15560264817486, 1.4626206819470515, 1.0, 1.2178482240718664, 1.6262032582174455, 1.0), (3.3341978045700875, 2.213176031893565, 1.0, 3.388324392306615, 2.5114736709748717, 1.0), (3.3341978045700875, 2.213176031893565, 1.0, 3.388324392306615, 2.5114736709748717, 1.0), (3.3341978045700875, 2.213176031893565, 1.0, 3.388324392306615, 2.5018511664883776, 1.0), (3.401856039240747, 2.40562612162344, 1.0, 3.461395285750927, 2.7712812921102032, 1.0), (3.3991497098539205, 2.40562612162344, 1.0, 3.461395285750927, 2.780903796596697, 1.0), (3.3964433804670944, 2.40562612162344, 1.0, 3.461395285750927, 2.780903796596697, 1.0), (3.3964433804670944, 2.40562612162344, 1.0, 3.461395285750927, 2.780903796596697, 1.0), (3.3964433804670944, 2.40562612162344, 1.0, 3.461395285750927, 2.780903796596697, 1.0), (3.3964433804670944, 2.40562612162344, 1.0, 3.461395285750927, 2.780903796596697, 1.0), (3.3639674278251777, 2.4441161395694153, 1.0, 3.445157309429969, 2.809771310056178, 1.0), (3.3639674278251777, 2.4441161395694153, 1.0, 3.445157309429969, 2.814582562299425, 1.0), (3.3612610984383515, 2.453738644055909, 1.0, 3.4424509800431426, 2.819393814542672, 1.0), (3.3612610984383515, 2.453738644055909, 1.0, 3.4424509800431426, 2.819393814542672, 1.0), (3.3612610984383515, 2.4441161395694153, 1.0, 3.4316256624958372, 2.8530725802454002, 1.0), (3.3612610984383515, 2.4441161395694153, 1.0, 3.4316256624958372, 2.857883832488647, 1.0), (3.3612610984383515, 2.4441161395694153, 1.0, 3.4316256624958372, 2.857883832488647, 1.0), (3.4153876861748786, 2.1891197706773307, 1.0, 3.461395285750927, 2.4729836530288964, 1.0), (3.4153876861748786, 2.1891197706773307, 1.0, 3.461395285750927, 2.4729836530288964, 1.0), (3.4153876861748786, 2.1891197706773307, 1.0, 3.461395285750927, 2.4729836530288964, 1.0), (3.4153876861748786, 2.1891197706773307, 1.0, 3.461395285750927, 2.4681724007856496, 1.0), (3.4153876861748786, 2.1891197706773307, 1.0, 3.461395285750927, 2.4633611485424027, 1.0), (3.4153876861748786, 2.1891197706773307, 1.0, 3.461395285750927, 2.4633611485424027, 1.0), (3.420800344948532, 2.184308518434084, 1.0, 3.461395285750927, 2.458549896299156, 1.0), (3.420800344948532, 2.184308518434084, 1.0, 3.461395285750927, 2.458549896299156, 1.0), (3.420800344948532, 2.184308518434084, 1.0, 3.461395285750927, 2.458549896299156, 1.0), (3.420800344948532, 2.184308518434084, 1.0, 3.461395285750927, 2.453738644055909, 1.0), (3.423506674335358, 2.184308518434084, 1.0, 3.461395285750927, 2.453738644055909, 1.0), (3.423506674335358, 2.184308518434084, 1.0, 3.461395285750927, 2.453738644055909, 1.0), (3.423506674335358, 2.1891197706773307, 1.0, 3.461395285750927, 2.453738644055909, 1.0), (3.423506674335358, 2.1891197706773307, 1.0, 3.461395285750927, 2.453738644055909, 1.0), (3.423506674335358, 2.184308518434084, 1.0, 3.461395285750927, 2.453738644055909, 1.0), (3.4262130037221845, 2.184308518434084, 1.0, 3.461395285750927, 2.453738644055909, 1.0), (3.423506674335358, 2.184308518434084, 1.0, 3.461395285750927, 2.453738644055909, 1.0), (3.4262130037221845, 2.184308518434084, 1.0, 3.461395285750927, 2.448927391812662, 1.0), (3.4289193331090106, 2.184308518434084, 1.0, 3.461395285750927, 2.448927391812662, 1.0), (3.4289193331090106, 2.1939310229205775, 1.0, 3.461395285750927, 2.448927391812662, 1.0), (3.4289193331090106, 2.1939310229205775, 1.0, 3.461395285750927, 2.448927391812662, 1.0), (3.3802054041461362, 2.208364779650318, 1.0, 3.4289193331090106, 2.492228662001884, 1.0), (3.3829117335329624, 2.208364779650318, 1.0, 3.4316256624958372, 2.492228662001884, 1.0), (3.3829117335329624, 2.213176031893565, 1.0, 3.4316256624958372, 2.492228662001884, 1.0), (3.3802054041461362, 2.208364779650318, 1.0, 3.4343319918826634, 2.492228662001884, 1.0), (3.3774990747593097, 2.203553527407071, 1.0, 3.4316256624958372, 2.492228662001884, 1.0), (3.3774990747593097, 2.203553527407071, 1.0, 3.43703832126949, 2.492228662001884, 1.0), (3.3774990747593097, 2.203553527407071, 1.0, 3.43703832126949, 2.492228662001884, 1.0), (3.3802054041461362, 2.1939310229205775, 1.0, 3.4343319918826634, 2.4633611485424027, 1.0), (3.3802054041461362, 2.1939310229205775, 1.0, 3.4343319918826634, 2.4633611485424027, 1.0), (2.619726846447926, 2.102517230298887, 1.0, 2.64679014031619, 2.256477302082787, 1.0), (2.6143141876742733, 2.1073284825421337, 1.0, 2.64679014031619, 2.256477302082787, 1.0), (2.6143141876742733, 2.1073284825421337, 1.0, 2.64679014031619, 2.256477302082787, 1.0), (2.6089015289006205, 2.09770597805564, 1.0, 2.64679014031619, 2.261288554326034, 1.0), (2.6089015289006205, 2.0880834735691463, 1.0, 2.64679014031619, 2.266099806569281, 1.0), (2.6089015289006205, 2.092894725812393, 1.0, 2.64679014031619, 2.256477302082787, 1.0), (2.6089015289006205, 2.0880834735691463, 1.0, 2.64679014031619, 2.256477302082787, 1.0), (2.6089015289006205, 2.0832722213258994, 1.0, 2.641377481542537, 2.256477302082787, 1.0), (2.6061951995137944, 2.078460969082652, 1.0, 2.6386711521557107, 2.266099806569281, 1.0), (2.6061951995137944, 2.078460969082652, 1.0, 2.635964822768884, 2.266099806569281, 1.0), (2.6061951995137944, 2.078460969082652, 1.0, 2.635964822768884, 2.266099806569281, 1.0), (1.2719748118083938, 2.208364779650318, 1.0, 1.3152760819976157, 2.424871130596428, 1.0), (1.266562153034741, 2.217987284136812, 1.0, 1.3152760819976157, 2.4200598783531806, 1.0), (1.2394988591664775, 2.1939310229205775, 1.0, 1.280093799968873, 2.40562612162344, 1.0), (1.2394988591664775, 2.1891197706773307, 1.0, 1.2773874705820467, 2.40562612162344, 1.0), (1.2340862003928246, 2.1939310229205775, 1.0, 1.2746811411952201, 2.40562612162344, 1.0), (1.2340862003928246, 2.184308518434084, 1.0, 1.2719748118083938, 2.40562612162344, 1.0), (1.2340862003928246, 2.213176031893565, 1.0, 1.2719748118083938, 2.400814869380193, 1.0), (1.207022906524561, 2.1698747617043432, 1.0, 1.2394988591664775, 2.3863811126504526, 1.0), (1.2043165771377347, 2.160252257217849, 1.0, 1.2422051885533039, 2.3960036171369463, 1.0), (1.2043165771377347, 2.17468601394759, 1.0, 1.2394988591664775, 2.3863811126504526, 1.0), (1.2016102477509083, 2.1554410049746022, 1.0, 1.2394988591664775, 2.3863811126504526, 1.0), (1.1934912595904292, 2.179497266190837, 1.0, 1.2367925297796512, 2.3815698604072058, 1.0), (1.1907849302036029, 2.179497266190837, 1.0, 1.2367925297796512, 2.376758608163959, 1.0), (1.1907849302036029, 2.1891197706773307, 1.0, 1.2340862003928246, 2.376758608163959, 1.0), (1.1907849302036029, 2.1891197706773307, 1.0, 1.2340862003928246, 2.376758608163959, 1.0), (1.18537227142995, 2.1939310229205775, 1.0, 1.2313798710059982, 2.371947355920712, 1.0), (1.18537227142995, 2.184308518434084, 1.0, 1.2313798710059982, 2.371947355920712, 1.0), (1.18537227142995, 2.1698747617043432, 1.0, 1.2286735416191719, 2.371947355920712, 1.0), (1.18537227142995, 2.17468601394759, 1.0, 1.2259672122323455, 2.371947355920712, 1.0), (1.1799596126562972, 2.1506297527313554, 1.0, 1.2205545534586928, 2.371947355920712, 1.0), (1.1799596126562972, 2.1506297527313554, 1.0, 1.2205545534586928, 2.371947355920712, 1.0), (1.1772532832694709, 2.1458185004881085, 1.0, 1.2178482240718664, 2.3575135991909715, 1.0), (1.1745469538826445, 2.1458185004881085, 1.0, 1.21514189468504, 2.3575135991909715, 1.0), (1.1745469538826445, 2.136195996001615, 1.0, 1.2124355652982137, 2.3623248514342183, 1.0), (1.1745469538826445, 2.136195996001615, 1.0, 1.2124355652982137, 2.3575135991909715, 1.0), (1.1745469538826445, 2.1410072482448617, 1.0, 1.2097292359113874, 2.3575135991909715, 1.0), (1.1718406244958182, 2.1458185004881085, 1.0, 1.207022906524561, 2.3478910947044778, 1.0), (1.1691342951089918, 2.1458185004881085, 1.0, 1.2043165771377347, 2.3478910947044778, 1.0), (1.163721636335339, 2.1554410049746022, 1.0, 1.2016102477509083, 2.3478910947044778, 1.0), (1.163721636335339, 2.165063509461096, 1.0, 1.2016102477509083, 2.3478910947044778, 1.0), (1.163721636335339, 2.165063509461096, 1.0, 1.198903918364082, 2.3478910947044778, 1.0), (1.163721636335339, 2.165063509461096, 1.0, 1.198903918364082, 2.3478910947044778, 1.0), (1.163721636335339, 2.165063509461096, 1.0, 1.1961975889772556, 2.343079842461231, 1.0), (1.1610153069485127, 2.165063509461096, 1.0, 1.1961975889772556, 2.3478910947044778, 1.0), (1.1610153069485127, 2.1554410049746022, 1.0, 1.1961975889772556, 2.3478910947044778, 1.0), (1.1583089775616864, 2.136195996001615, 1.0, 1.2016102477509083, 2.3478910947044778, 1.0), (1.1528963187880337, 2.131384743758368, 1.0, 1.198903918364082, 2.3478910947044778, 1.0), (1.1528963187880337, 2.131384743758368, 1.0, 1.1961975889772556, 2.338268590217984, 1.0), (1.1528963187880337, 2.131384743758368, 1.0, 1.1961975889772556, 2.32864608573149, 1.0), (1.1501899894012073, 2.131384743758368, 1.0, 1.1934912595904292, 2.32864608573149, 1.0), (1.1501899894012073, 2.131384743758368, 1.0, 1.1934912595904292, 2.3334573379747368, 1.0), (1.1501899894012073, 2.126573491515121, 1.0, 1.1907849302036029, 2.32864608573149, 1.0), (1.2692684824215674, 2.213176031893565, 1.0, 1.3152760819976157, 2.424871130596428, 1.0), (1.266562153034741, 2.213176031893565, 1.0, 1.3152760819976157, 2.424871130596428, 1.0), (1.2611494942610884, 2.1891197706773307, 1.0, 1.3152760819976157, 2.424871130596428, 1.0), (1.258443164874262, 2.1939310229205775, 1.0, 1.3152760819976157, 2.424871130596428, 1.0), (1.258443164874262, 2.1939310229205775, 1.0, 1.3152760819976157, 2.424871130596428, 1.0), (1.2367925297796512, 2.184308518434084, 1.0, 1.2773874705820467, 2.40562612162344, 1.0), (1.2340862003928246, 2.184308518434084, 1.0, 1.2773874705820467, 2.410437373866687, 1.0), (1.2286735416191719, 2.165063509461096, 1.0, 1.2719748118083938, 2.40562612162344, 1.0), (1.2286735416191719, 2.165063509461096, 1.0, 1.2719748118083938, 2.40562612162344, 1.0), (1.2016102477509083, 2.160252257217849, 1.0, 1.2394988591664775, 2.3863811126504526, 1.0), (1.198903918364082, 2.160252257217849, 1.0, 1.2394988591664775, 2.3815698604072058, 1.0), (1.198903918364082, 2.160252257217849, 1.0, 1.2394988591664775, 2.3815698604072058, 1.0), (3.3396104633437407, 2.213176031893565, 1.0, 3.393737051080268, 2.4970399142451307, 1.0), (3.3396104633437407, 2.213176031893565, 1.0, 3.393737051080268, 2.4970399142451307, 1.0), (3.3639674278251777, 2.1891197706773307, 1.0, 3.4180940155617052, 2.453738644055909, 1.0), (3.3639674278251777, 2.184308518434084, 1.0, 3.4180940155617052, 2.453738644055909, 1.0), (3.3639674278251777, 2.184308518434084, 1.0, 3.4180940155617052, 2.453738644055909, 1.0), (3.3639674278251777, 2.184308518434084, 1.0, 3.4180940155617052, 2.448927391812662, 1.0), (3.3639674278251777, 2.184308518434084, 1.0, 3.4180940155617052, 2.448927391812662, 1.0), (3.3639674278251777, 2.1891197706773307, 1.0, 3.4180940155617052, 2.448927391812662, 1.0), (3.3666737572120042, 2.1939310229205775, 1.0, 3.420800344948532, 2.4441161395694153, 1.0), (3.3666737572120042, 2.1891197706773307, 1.0, 3.420800344948532, 2.448927391812662, 1.0), (3.3666737572120042, 2.1891197706773307, 1.0, 3.420800344948532, 2.448927391812662, 1.0), (3.3666737572120042, 2.184308518434084, 1.0, 3.420800344948532, 2.453738644055909, 1.0), (3.3666737572120042, 2.184308518434084, 1.0, 3.420800344948532, 2.453738644055909, 1.0), (3.3666737572120042, 2.184308518434084, 1.0, 3.420800344948532, 2.448927391812662, 1.0), (3.3666737572120042, 2.184308518434084, 1.0, 3.420800344948532, 2.448927391812662, 1.0), (3.3666737572120042, 2.184308518434084, 1.0, 3.420800344948532, 2.4441161395694153, 1.0), (3.3666737572120042, 2.184308518434084, 1.0, 3.420800344948532, 2.4441161395694153, 1.0), (3.3666737572120042, 2.184308518434084, 1.0, 3.420800344948532, 2.4441161395694153, 1.0), (3.3693800865988304, 2.1891197706773307, 1.0, 3.4180940155617052, 2.4393048873261685, 1.0), (3.3666737572120042, 2.184308518434084, 1.0, 3.420800344948532, 2.4393048873261685, 1.0), (3.374792745372483, 2.1891197706773307, 1.0, 3.4180940155617052, 2.4441161395694153, 1.0), (3.3774990747593097, 2.184308518434084, 1.0, 3.420800344948532, 2.4441161395694153, 1.0), (3.3774990747593097, 2.184308518434084, 1.0, 3.420800344948532, 2.4441161395694153, 1.0), (3.3829117335329624, 2.1891197706773307, 1.0, 3.4316256624958372, 2.4393048873261685, 1.0), (3.3774990747593097, 2.1891197706773307, 1.0, 3.43703832126949, 2.4344936350829216, 1.0), (3.3774990747593097, 2.1939310229205775, 1.0, 3.43703832126949, 2.4344936350829216, 1.0), (3.3802054041461362, 2.1939310229205775, 1.0, 3.4343319918826634, 2.4344936350829216, 1.0), (3.3829117335329624, 2.1891197706773307, 1.0, 3.4316256624958372, 2.4393048873261685, 1.0), (3.3829117335329624, 2.1891197706773307, 1.0, 3.4316256624958372, 2.4344936350829216, 1.0), (3.385618062919789, 2.1939310229205775, 1.0, 3.4289193331090106, 2.4344936350829216, 1.0), (3.388324392306615, 2.1891197706773307, 1.0, 3.4262130037221845, 2.4344936350829216, 1.0), (3.388324392306615, 2.184308518434084, 1.0, 3.4262130037221845, 2.4296823828396747, 1.0), (3.388324392306615, 2.184308518434084, 1.0, 3.4262130037221845, 2.4296823828396747, 1.0), (3.385618062919789, 2.184308518434084, 1.0, 3.4289193331090106, 2.424871130596428, 1.0), (3.385618062919789, 2.179497266190837, 1.0, 3.4289193331090106, 2.424871130596428, 1.0), (3.385618062919789, 2.184308518434084, 1.0, 3.4289193331090106, 2.424871130596428, 1.0), (3.388324392306615, 2.179497266190837, 1.0, 3.4316256624958372, 2.4200598783531806, 1.0), (3.385618062919789, 2.179497266190837, 1.0, 3.4343319918826634, 2.4152486261099337, 1.0), (3.3910307216934417, 2.179497266190837, 1.0, 3.4343319918826634, 2.4152486261099337, 1.0), (3.3964433804670944, 2.184308518434084, 1.0, 3.439744650656316, 2.4152486261099337, 1.0), (3.3964433804670944, 2.179497266190837, 1.0, 3.445157309429969, 2.4152486261099337, 1.0), (3.3964433804670944, 2.179497266190837, 1.0, 3.445157309429969, 2.4152486261099337, 1.0), (2.116349580498221, 3.2331615074619036, 1.0, 2.7117420456000225, 3.4544791106512602, 1.0), (2.116349580498221, 3.165803976056447, 1.0, 2.700916728052717, 3.4544791106512602, 1.0), (2.1190559098850477, 3.1561814715699534, 1.0, 2.698210398665891, 3.4544791106512602, 1.0), (2.129881227432353, 3.1273139581104723, 1.0, 2.6927977398922383, 3.4544791106512602, 1.0), (2.135293886206006, 3.088823940164497, 1.0, 2.6873850811185855, 3.4496678584080134, 1.0), (2.143412874366485, 3.059956426705016, 1.0, 2.684678751731759, 3.4496678584080134, 1.0), (2.1380002155928324, 3.0407114177320285, 1.0, 2.6900914105054117, 3.4544791106512602, 1.0), (2.146119203753311, 3.021466408759041, 1.0, 2.6792660929581062, 3.4544791106512602, 1.0), (2.1596508506874432, 2.978165138569819, 1.0, 2.668440775410801, 3.4496678584080134, 1.0), (2.165063509461096, 2.934863868380597, 1.0, 2.663028116637148, 3.4544791106512602, 1.0), (2.1758888270084014, 2.9059963549211156, 1.0, 2.64679014031619, 3.4544791106512602, 1.0), (2.200245791489839, 2.857883832488647, 1.0, 2.6089015289006205, 3.459290362894507, 1.0), (2.208364779650318, 2.8482613280021534, 1.0, 2.6007825407401417, 3.459290362894507, 1.0), (2.2191900971976235, 2.838638823515659, 1.0, 2.5926635525796624, 3.4544791106512602, 1.0), (2.213777438423971, 2.814582562299425, 1.0, 2.5872508938060097, 3.4448566061647665, 1.0), (2.2191900971976235, 2.8049600578129312, 1.0, 2.5764255762587043, 3.4352341016782724, 1.0), (2.208364779650318, 2.7905263010831907, 1.0, 2.5872508938060097, 3.4544791106512602, 1.0), (2.208364779650318, 2.77609254435345, 1.0, 2.581838235032357, 3.459290362894507, 1.0), (2.208364779650318, 2.7712812921102032, 1.0, 2.5764255762587043, 3.4544791106512602, 1.0), (2.208364779650318, 2.7616587876237095, 1.0, 2.5710129174850516, 3.4544791106512602, 1.0), (2.211071109037144, 2.7472250308939685, 1.0, 2.568306588098225, 3.4544791106512602, 1.0), (2.213777438423971, 2.737602526407475, 1.0, 2.565600258711399, 3.4544791106512602, 1.0), (2.2191900971976235, 2.732791274164228, 1.0, 2.565600258711399, 3.4352341016782724, 1.0), (2.2191900971976235, 2.7135462651912405, 1.0, 2.565600258711399, 3.415989092705285, 1.0), (2.2191900971976235, 2.694301256218253, 1.0, 2.565600258711399, 3.3967440837322975, 1.0), (2.224602755971276, 2.6798674994885125, 1.0, 2.5710129174850516, 3.37749907475931, 1.0), (2.230015414744929, 2.6750562472452653, 1.0, 2.581838235032357, 3.3726878225160632, 1.0), (2.2381344029054078, 2.6750562472452653, 1.0, 2.589957223192836, 3.339009056813335, 1.0), (2.2435470616790605, 2.6654337427587715, 1.0, 2.598076211353315, 3.3197640478403474, 1.0), (2.248959720452713, 2.6654337427587715, 1.0, 2.6089015289006205, 3.295707786624113, 1.0), (2.2570787086131925, 2.646188733785784, 1.0, 2.619726846447926, 3.2812740298943726, 1.0), (2.259785038000019, 2.6317549770560436, 1.0, 2.633258493382058, 3.257217768678138, 1.0), (2.267904026160498, 2.62213247256955, 1.0, 2.64679014031619, 3.2331615074619036, 1.0), (2.2787293437078033, 2.6076987158398093, 1.0, 2.663028116637148, 3.213916498488916, 1.0), (2.284142002481456, 2.5980762113533156, 1.0, 2.668440775410801, 3.1946714895159287, 1.0), (2.2949673200287615, 2.5884537068668214, 1.0, 2.6792660929581062, 3.165803976056447, 1.0), (2.3057926375760673, 2.574019950137081, 1.0, 2.6900914105054117, 3.1465589670834597, 1.0), (2.3247369432838516, 2.574019950137081, 1.0, 2.7036230574395437, 3.1176914536239786, 1.0), (2.3382685902179836, 2.5595861934073403, 1.0, 2.7117420456000225, 3.0984464446509907, 1.0), (2.349093907765289, 2.5547749411640934, 1.0, 2.727980021920981, 3.0743901834347565, 1.0), (2.3599192253125945, 2.540341184434353, 1.0, 2.7415116688551127, 3.050333922218522, 1.0), (2.3761572016335526, 2.530718679947859, 1.0, 2.749630657015592, 3.0310889132455348, 1.0), (2.3869825191808585, 2.5162849232181186, 1.0, 2.7604559745628974, 3.016655156515794, 1.0), (2.4059268248886427, 2.506662418731625, 1.0, 2.7739876214970294, 2.9974101475428063, 1.0), (2.4194584718227747, 2.4970399142451307, 1.0, 2.7902255978179875, 2.978165138569819, 1.0), (2.43028378937008, 2.48260615751539, 1.0, 2.803757244752119, 2.9541088773535846, 1.0), (2.4465217656910383, 2.4777949052721433, 1.0, 2.8199952210730777, 2.939675120623844, 1.0), (2.4627597420119964, 2.4633611485424027, 1.0, 2.833526868007209, 2.9156188594076093, 1.0), (2.478997718332955, 2.458549896299156, 1.0, 2.847058514941341, 2.9011851026778688, 1.0), (2.4898230358802604, 2.4441161395694153, 1.0, 2.847058514941341, 2.896373850434622, 1.0), (2.5060610122012186, 2.4344936350829216, 1.0, 2.868709150035952, 2.867506336975141, 1.0), (2.5195926591353506, 2.4344936350829216, 1.0, 2.8795344675832575, 2.857883832488647, 1.0), (2.533124306069482, 2.4296823828396747, 1.0, 2.88494712635691, 2.843450075758906, 1.0), (2.546655953003614, 2.424871130596428, 1.0, 2.8957724439042156, 2.838638823515659, 1.0), (2.560187599937746, 2.4152486261099337, 1.0, 2.912010420225174, 2.819393814542672, 1.0), (2.5710129174850516, 2.4152486261099337, 1.0, 2.9228357377724796, 2.809771310056178, 1.0), (2.584544564419183, 2.400814869380193, 1.0, 2.9282483965461323, 2.8001488055696844, 1.0), (2.598076211353315, 2.3911923648936995, 1.0, 2.936367384706611, 2.785715048839944, 1.0), (2.603488870126968, 2.3863811126504526, 1.0, 2.9444863728670905, 2.77609254435345, 1.0), (2.6143141876742733, 2.3863811126504526, 1.0, 2.947192702253917, 2.7616587876237095, 1.0), (2.6251395052215787, 2.3863811126504526, 1.0, 2.9526053610275698, 2.7568475353804622, 1.0), (2.641377481542537, 2.376758608163959, 1.0, 2.9607243491880486, 2.7424137786507217, 1.0), (2.6440838109293634, 2.367136103677465, 1.0, 2.9742559961221806, 2.732791274164228, 1.0), (2.654909128476669, 2.3623248514342183, 1.0, 2.9796686548958333, 2.7231687696777342, 1.0), (2.668440775410801, 2.3575135991909715, 1.0, 2.987787643056312, 2.7135462651912405, 1.0), (2.6792660929581062, 2.3575135991909715, 1.0, 2.9932003018299653, 2.7135462651912405, 1.0), (2.684678751731759, 2.3575135991909715, 1.0, 2.998612960603618, 2.7135462651912405, 1.0), (2.6955040692790644, 2.3527023469477246, 1.0, 3.0094382781509235, 2.6894900039750063, 1.0), (2.7036230574395437, 2.3478910947044778, 1.0, 3.0121446075377496, 2.6846787517317594, 1.0), (2.7117420456000225, 2.338268590217984, 1.0, 3.020263595698229, 2.6750562472452653, 1.0), (2.7171547043736752, 2.32864608573149, 1.0, 3.0283825838587077, 2.6750562472452653, 1.0), (2.733392680694634, 2.319023581244996, 1.0, 3.036501572019187, 2.6654337427587715, 1.0), (2.7442179982419392, 2.319023581244996, 1.0, 3.0446205601796663, 2.655811238272278, 1.0), (2.749630657015592, 2.319023581244996, 1.0, 3.0473268895664924, 2.646188733785784, 1.0), (2.7631623039497235, 2.3142123290017493, 1.0, 3.0608585365006244, 2.6413774815425373, 1.0), (2.7739876214970294, 2.3094010767585025, 1.0, 3.066271195274277, 2.6365662292992904, 1.0), (2.779400280270682, 2.3094010767585025, 1.0, 3.07168385404793, 2.6269437248127967, 1.0), (2.7902255978179875, 2.2997785722720088, 1.0, 3.0770965128215826, 2.617321220326303, 1.0), (2.7983445859784664, 2.2997785722720088, 1.0, 3.085215500982062, 2.612509968083056, 1.0), (2.809169903525772, 2.294967320028762, 1.0, 3.0960408185293673, 2.6028874635965624, 1.0), (2.533124306069482, 2.1169509870286274, 1.0, 2.5764255762587043, 2.319023581244996, 1.0), (2.533124306069482, 2.1169509870286274, 1.0, 2.5764255762587043, 2.319023581244996, 1.0), (2.514180000361698, 2.1121397347853805, 1.0, 2.5628939293245723, 2.32864608573149, 1.0), (2.514180000361698, 2.1121397347853805, 1.0, 2.5628939293245723, 2.32864608573149, 1.0), (2.5060610122012186, 2.1217622392718742, 1.0, 2.538536964843135, 2.323834833488243, 1.0), (2.503354682814392, 2.1121397347853805, 1.0, 2.5412432942299614, 2.32864608573149, 1.0), (2.503354682814392, 2.1169509870286274, 1.0, 2.5412432942299614, 2.3334573379747368, 1.0), (2.500648353427566, 2.1121397347853805, 1.0, 2.538536964843135, 2.3334573379747368, 1.0), (2.4979420240407393, 2.1073284825421337, 1.0, 2.538536964843135, 2.3334573379747368, 1.0), (2.4979420240407393, 2.1073284825421337, 1.0, 2.538536964843135, 2.338268590217984, 1.0), (2.4979420240407393, 2.1121397347853805, 1.0, 2.538536964843135, 2.338268590217984, 1.0), (2.4898230358802604, 2.1217622392718742, 1.0, 2.538536964843135, 2.3334573379747368, 1.0), (2.4898230358802604, 2.1217622392718742, 1.0, 2.538536964843135, 2.3334573379747368, 1.0), (2.9255420671593058, 1.967802167487974, 1.0, 2.968843337348528, 2.1169509870286274, 1.0), (2.9309547259329585, 1.9629909152447271, 1.0, 2.968843337348528, 2.1169509870286274, 1.0), (2.9309547259329585, 1.9629909152447271, 1.0, 2.968843337348528, 2.1169509870286274, 1.0), (2.414045813049122, 2.1458185004881085, 1.0, 2.4627597420119964, 2.376758608163959, 1.0), (2.414045813049122, 2.1458185004881085, 1.0, 2.4627597420119964, 2.376758608163959, 1.0), (2.9823749842826595, 1.9726134197312208, 1.0, 3.0040256193772708, 2.102517230298887, 1.0), (2.9823749842826595, 1.9726134197312208, 1.0, 3.0040256193772708, 2.102517230298887, 1.0), (2.3247369432838516, 2.17468601394759, 1.0, 2.3815698604072058, 2.3960036171369463, 1.0), (2.3220306138970255, 2.17468601394759, 1.0, 2.3761572016335526, 2.400814869380193, 1.0), (2.3220306138970255, 2.17468601394759, 1.0, 2.3761572016335526, 2.400814869380193, 1.0), (2.181301485782054, 2.857883832488647, 1.0, 2.6386711521557107, 3.4448566061647665, 1.0), (0.446544348826351, 1.934123401785246, 1.0, 0.5304405598179684, 2.0255371944069367, 1.0), (0.446544348826351, 1.934123401785246, 1.0, 0.5304405598179684, 2.0255371944069367, 1.0), (0.5087899247233576, 1.934123401785246, 1.0, 0.5845671475544959, 2.0303484466501835, 1.0), (0.5087899247233576, 1.934123401785246, 1.0, 0.5845671475544959, 2.0303484466501835, 1.0), (0.41948105495808735, 1.9148783928122584, 1.0, 0.511496254110184, 2.02072594216369, 1.0), (0.41948105495808735, 1.9148783928122584, 1.0, 0.511496254110184, 2.02072594216369, 1.0), (0.3978304198634764, 1.924500897298752, 1.0, 0.49255194840239935, 2.02072594216369, 1.0), (0.3978304198634764, 1.924500897298752, 1.0, 0.49255194840239935, 2.02072594216369, 1.0), (2.70632938682637, 2.0880834735691463, 1.0, 2.7388053394682865, 2.2324210408665524, 1.0), (2.70632938682637, 2.0880834735691463, 1.0, 2.7388053394682865, 2.2324210408665524, 1.0), (2.641377481542537, 2.0832722213258994, 1.0, 2.6792660929581062, 2.256477302082787, 1.0), (2.641377481542537, 2.0880834735691463, 1.0, 2.6792660929581062, 2.261288554326034, 1.0), (2.641377481542537, 2.0832722213258994, 1.0, 2.6792660929581062, 2.25166604983954, 1.0), (2.641377481542537, 2.0832722213258994, 1.0, 2.6738534341844535, 2.25166604983954, 1.0), (2.6386711521557107, 2.078460969082652, 1.0, 2.671147104797627, 2.242043545353046, 1.0), (2.6386711521557107, 2.078460969082652, 1.0, 2.668440775410801, 2.25166604983954, 1.0), (2.635964822768884, 2.0880834735691463, 1.0, 2.6738534341844535, 2.242043545353046, 1.0), (2.635964822768884, 2.078460969082652, 1.0, 2.6657344460239742, 2.256477302082787, 1.0), (2.635964822768884, 2.078460969082652, 1.0, 2.6657344460239742, 2.256477302082787, 1.0), (3.3612610984383515, 2.203553527407071, 1.0, 3.409975027401226, 2.4441161395694153, 1.0), (3.3612610984383515, 2.203553527407071, 1.0, 3.409975027401226, 2.4441161395694153, 1.0), (3.3396104633437407, 2.4441161395694153, 1.0, 3.420800344948532, 2.886751345948128, 1.0), (3.3314914751832614, 2.4633611485424027, 1.0, 3.423506674335358, 2.8771288414616345, 1.0), (3.3314914751832614, 2.458549896299156, 1.0, 3.423506674335358, 2.8771288414616345, 1.0), (3.3314914751832614, 2.4633611485424027, 1.0, 3.423506674335358, 2.8771288414616345, 1.0), (3.3314914751832614, 2.458549896299156, 1.0, 3.4180940155617052, 2.886751345948128, 1.0), (3.3341978045700875, 2.4633611485424027, 1.0, 3.4153876861748786, 2.896373850434622, 1.0), (3.3341978045700875, 2.458549896299156, 1.0, 3.4153876861748786, 2.9059963549211156, 1.0), (3.3314914751832614, 2.453738644055909, 1.0, 3.4126813567880525, 2.9156188594076093, 1.0), (3.328785145796435, 2.453738644055909, 1.0, 3.4153876861748786, 2.9156188594076093, 1.0), (3.3260788164096087, 2.4633611485424027, 1.0, 3.4126813567880525, 2.925241363894103, 1.0), (3.323372487022782, 2.4681724007856496, 1.0, 3.409975027401226, 2.9156188594076093, 1.0), (3.320666157635956, 2.4729836530288964, 1.0, 3.4072686980144, 2.9156188594076093, 1.0), (3.3179598282491294, 2.4729836530288964, 1.0, 3.404562368627573, 2.9156188594076093, 1.0), (3.3179598282491294, 2.4777949052721433, 1.0, 3.3991497098539205, 2.9059963549211156, 1.0), (3.3152534988623032, 2.4729836530288964, 1.0, 3.3964433804670944, 2.9156188594076093, 1.0), (3.3125471694754767, 2.4729836530288964, 1.0, 3.3991497098539205, 2.9156188594076093, 1.0), (3.3098408400886505, 2.48260615751539, 1.0, 3.3964433804670944, 2.9156188594076093, 1.0), (3.3125471694754767, 2.48260615751539, 1.0, 3.393737051080268, 2.920430111650856, 1.0), (3.3098408400886505, 2.487417409758637, 1.0, 3.3910307216934417, 2.925241363894103, 1.0), (3.3098408400886505, 2.492228662001884, 1.0, 3.3910307216934417, 2.934863868380597, 1.0), (3.3098408400886505, 2.492228662001884, 1.0, 3.3910307216934417, 2.934863868380597, 1.0), (3.304428181314998, 2.492228662001884, 1.0, 3.3910307216934417, 2.939675120623844, 1.0), (3.304428181314998, 2.492228662001884, 1.0, 3.3910307216934417, 2.934863868380597, 1.0), (3.307134510701824, 2.492228662001884, 1.0, 3.388324392306615, 2.944486372867091, 1.0), (3.3098408400886505, 2.492228662001884, 1.0, 3.3910307216934417, 2.9541088773535846, 1.0), (3.3098408400886505, 2.492228662001884, 1.0, 3.3910307216934417, 2.9637313818400783, 1.0), (3.307134510701824, 2.5018511664883776, 1.0, 3.393737051080268, 2.973353886326572, 1.0), (3.304428181314998, 2.506662418731625, 1.0, 3.3910307216934417, 2.973353886326572, 1.0), (3.304428181314998, 2.5114736709748717, 1.0, 3.3910307216934417, 2.973353886326572, 1.0), (3.299015522541345, 2.5162849232181186, 1.0, 3.3910307216934417, 2.978165138569819, 1.0), (3.2963091931545185, 2.5162849232181186, 1.0, 3.388324392306615, 2.978165138569819, 1.0), (3.2936028637676924, 2.5210961754613654, 1.0, 3.385618062919789, 2.978165138569819, 1.0), (3.290896534380866, 2.5259074277046123, 1.0, 3.3829117335329624, 2.978165138569819, 1.0), (3.2881902049940392, 2.530718679947859, 1.0, 3.3802054041461362, 2.973353886326572, 1.0), (3.2881902049940392, 2.5210961754613654, 1.0, 3.374792745372483, 2.9829763908130658, 1.0), (3.285483875607213, 2.5259074277046123, 1.0, 3.372086415985657, 2.9974101475428063, 1.0), (3.285483875607213, 2.530718679947859, 1.0, 3.372086415985657, 2.9925988952995595, 1.0), (3.285483875607213, 2.530718679947859, 1.0, 3.372086415985657, 2.9925988952995595, 1.0), (3.285483875607213, 2.535529932191106, 1.0, 3.372086415985657, 2.9974101475428063, 1.0), (3.2827775462203865, 2.540341184434353, 1.0, 3.3693800865988304, 3.002221399786053, 1.0), (3.2800712168335604, 2.540341184434353, 1.0, 3.372086415985657, 3.0070326520293, 1.0), (3.277364887446734, 2.540341184434353, 1.0, 3.3693800865988304, 3.0070326520293, 1.0), (3.2746585580599077, 2.540341184434353, 1.0, 3.372086415985657, 3.0070326520293, 1.0), (1.3639900109604906, 3.08401268792125, 1.0, 2.165063509461096, 3.459290362894507, 1.0), (1.3369267170922268, 3.0407114177320285, 1.0, 2.1380002155928324, 3.4544791106512602, 1.0), (1.2530305061006093, 2.944486372867091, 1.0, 2.070341980922173, 3.4544791106512602, 1.0), (1.2530305061006093, 2.944486372867091, 1.0, 2.070341980922173, 3.4544791106512602, 1.0), (1.2232608828455191, 2.896373850434622, 1.0, 2.0080964050251664, 3.459290362894507, 1.0), (1.2097292359113874, 2.886751345948128, 1.0, 1.9675014642227708, 3.4448566061647665, 1.0), (1.2016102477509083, 2.857883832488647, 1.0, 1.953969817288639, 3.4448566061647665, 1.0), (1.1772532832694709, 2.838638823515659, 1.0, 1.9269065234203753, 3.4448566061647665, 1.0), (1.1610153069485127, 2.819393814542672, 1.0, 1.8917242413916326, 3.4544791106512602, 1.0), (1.15560264817486, 2.809771310056178, 1.0, 1.8673672769101952, 3.4544791106512602, 1.0), (1.1474836600143807, 2.8001488055696844, 1.0, 1.8484229712024107, 3.459290362894507, 1.0), (1.1366583424670753, 2.785715048839944, 1.0, 1.8186533479473206, 3.459290362894507, 1.0), (1.1285393543065962, 2.7520362831372154, 1.0, 1.7888837246922304, 3.459290362894507, 1.0), (1.1177140367592908, 2.7424137786507217, 1.0, 1.7537014426634876, 3.459290362894507, 1.0), (1.098769731051506, 2.732791274164228, 1.0, 1.7320508075688767, 3.459290362894507, 1.0), (1.0852380841173743, 2.7183575174344874, 1.0, 1.710400172474266, 3.459290362894507, 1.0), (1.0717064371832425, 2.694301256218253, 1.0, 1.6779242198323494, 3.459290362894507, 1.0), (1.0581747902491105, 2.6654337427587715, 1.0, 1.661686243511391, 3.4352341016782724, 1.0), (1.0662937784095898, 2.646188733785784, 1.0, 1.6346229496431275, 3.4063665882187912, 1.0), (1.060881119635937, 2.655811238272278, 1.0, 1.6210913027089955, 3.37749907475931, 1.0), (1.0473494727018051, 2.646188733785784, 1.0, 1.602146997001211, 3.339009056813335, 1.0), (1.0365241551544997, 2.62213247256955, 1.0, 1.5804963619066, 3.3101415433538537, 1.0), (1.031111496380847, 2.617321220326303, 1.0, 1.558845726811989, 3.2908965343808663, 1.0), (1.022992508220368, 2.5980762113533156, 1.0, 1.542607750491031, 3.2620290209213847, 1.0), (1.0148735200598886, 2.5884537068668214, 1.0, 1.5209571153964199, 3.2331615074619036, 1.0), (1.323395070158095, 3.0359001654887816, 1.0, 2.1380002155928324, 3.459290362894507, 1.0), (1.323395070158095, 3.0359001654887816, 1.0, 2.1380002155928324, 3.459290362894507, 1.0), (1.2638558236479147, 2.9637313818400783, 1.0, 2.0865799572431314, 3.459290362894507, 1.0)]\n" + ] + } + ], + "source": [ + "geos = filtered_world.get_bbox_geo()\n", + "print(\"Bbox geos are:\",geos)" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "a548cc6e", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Trajectory distances are:\n", + "[('{0@2021-06-08 07:10:30+00, 0@2021-06-08 07:10:31+00, 0@2021-06-08 07:10:32+00, 0@2021-06-08 07:10:33+00, 0@2021-06-08 07:10:34+00, 0@2021-06-08 07:10:35+00, 0@2021-06-08 07:10:36+00, 0@2021-06-08 07:10:37+00, 0@2021-06-08 07:10:38+00, 0@2021-06-08 07:10:39+00, 0@2021-06-08 07:10:40+00, 0@2021-06-08 07:10:41+00, 0@2021-06-08 07:10:42+00, 0@2021-06-08 07:10:43+00, 0@2021-06-08 07:10:44+00, 0@2021-06-08 07:10:45+00, 0@2021-06-08 07:10:46+00, 0@2021-06-08 07:10:47+00, 0@2021-06-08 07:10:48+00, 0@2021-06-08 07:10:49+00, 0@2021-06-08 07:10:50+00, 0@2021-06-08 07:10:51+00, 0@2021-06-08 07:10:52+00, 0@2021-06-08 07:10:53+00, 0@2021-06-08 07:10:54+00, 0@2021-06-08 07:10:55+00, 0@2021-06-08 07:10:56+00, 0@2021-06-08 07:10:57+00}',), ('{0@2021-06-08 07:10:30+00, 0@2021-06-08 07:10:31+00, 0@2021-06-08 07:10:32+00, 0@2021-06-08 07:10:33+00, 0@2021-06-08 07:10:34+00, 0@2021-06-08 07:10:35+00, 0@2021-06-08 07:10:36+00, 0@2021-06-08 07:10:37+00, 0@2021-06-08 07:10:38+00, 0@2021-06-08 07:10:39+00, 0@2021-06-08 07:10:40+00, 0@2021-06-08 07:10:41+00, 0@2021-06-08 07:10:42+00, 0@2021-06-08 07:10:43+00, 0@2021-06-08 07:10:44+00, 0@2021-06-08 07:10:45+00, 0@2021-06-08 07:10:46+00, 0@2021-06-08 07:10:47+00, 0@2021-06-08 07:10:48+00, 0@2021-06-08 07:10:49+00, 0@2021-06-08 07:10:50+00, 0@2021-06-08 07:10:51+00, 0@2021-06-08 07:10:52+00, 0@2021-06-08 07:10:53+00, 0@2021-06-08 07:10:54+00, 0@2021-06-08 07:10:55+00, 0@2021-06-08 07:10:56+00, 0@2021-06-08 07:10:57+00}',), ('{0@2021-06-08 07:10:37+00, 0@2021-06-08 07:10:38+00, 0@2021-06-08 07:10:39+00, 0@2021-06-08 07:10:40+00, 0@2021-06-08 07:10:41+00, 0@2021-06-08 07:10:42+00, 0@2021-06-08 07:10:43+00, 0@2021-06-08 07:10:44+00, 0@2021-06-08 07:10:45+00, 0@2021-06-08 07:10:46+00, 0@2021-06-08 07:10:47+00, 0@2021-06-08 07:10:49+00, 0@2021-06-08 07:10:50+00, 0@2021-06-08 07:10:51+00, 0@2021-06-08 07:10:52+00, 0@2021-06-08 07:10:53+00, 0@2021-06-08 07:10:54+00, 0@2021-06-08 07:10:55+00, 0@2021-06-08 07:10:56+00, 0@2021-06-08 07:10:57+00}',), (None,)]\n" + ] + } + ], + "source": [ + "print(\"Trajectory distances are:\")\n", + "print(filtered_world.get_distance(0, 30))" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "c20e8370", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Trajectory speeds are:\n", + "[(None,), (None,), (None,), (None,)]\n" + ] + } + ], + "source": [ + "print(\"Trajectory speeds are:\")\n", + "print(filtered_world.get_speed(0, 30))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8e1aa7af", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.8" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/scenic_equivelants.ipynb b/scenic_equivelants.ipynb new file mode 100644 index 00000000..f8175287 --- /dev/null +++ b/scenic_equivelants.ipynb @@ -0,0 +1,588 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 25, + "metadata": {}, + "outputs": [], + "source": [ + "# Deleting the .apperception_cache if it exists, as to avoid DB conflict errors\n", + "import os\n", + "import shutil\n", + "\n", + "dirpath = os.path.join('.apperception_cache')\n", + "if os.path.exists(dirpath) and os.path.isdir(dirpath):\n", + " shutil.rmtree(dirpath)\n", + "\n", + "dirpath = os.path.join('output')\n", + "if os.path.exists(dirpath) and os.path.isdir(dirpath):\n", + " shutil.rmtree(dirpath)\n", + "os.mkdir(dirpath)\n", + "\n", + "# This piece of code is unsafe, and should not be run if not needed. \n", + "# It serves for test purposes when one recieves a \"dead kernel\" error.\n", + "os.environ[\"KMP_DUPLICATE_LIB_OK\"]=\"TRUE\"" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": {}, + "outputs": [ + { + "ename": "FileNotFoundError", + "evalue": "[Errno 2] No such file or directory: '.apperception_cache/2022-03-04 07;30;49.195866_b0836101-4935-450b-a2e0-22f469cf8994_trafficScene.ap.yaml'", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mFileNotFoundError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m/var/folders/jk/04mdk5z529z3kdnl2cl6ty3h0000gn/T/ipykernel_22556/3245952245.py\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[1;32m 17\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 18\u001b[0m \u001b[0;31m# 1. define a world\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 19\u001b[0;31m \u001b[0mtraffic_world\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mempty_world\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 20\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 21\u001b[0m \u001b[0;31m# 2. construct a camera\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/Desktop/apperception/apperception/new_world.py\u001b[0m in \u001b[0;36mempty_world\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m 480\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmatched_files\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 481\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0m_empty_world_from_file\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmatched_files\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 482\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0m_empty_world\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 483\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 484\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/Desktop/apperception/apperception/new_world.py\u001b[0m in \u001b[0;36m_empty_world\u001b[0;34m(name)\u001b[0m\n\u001b[1;32m 495\u001b[0m \u001b[0mtimestamp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdatetime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdatetime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mutcnow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 496\u001b[0m \u001b[0mlog_file\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfilename\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtimestamp\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mworld_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 497\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlog_file\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"w\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 498\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0myaml\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msafe_dump\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 499\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mWorld\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mworld_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtimestamp\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: '.apperception_cache/2022-03-04 07;30;49.195866_b0836101-4935-450b-a2e0-22f469cf8994_trafficScene.ap.yaml'" + ] + } + ], + "source": [ + "import sys\n", + "sys.path.append(os.path.join(os.getcwd(),\"apperception\"))\n", + "\n", + "### IMPORTS\n", + "import lens\n", + "import point\n", + "from new_world import empty_world\n", + "\n", + "# Let's define some attribute for constructing the world first\n", + "name = \"trafficScene\" # world name\n", + "units = \"metrics\" # world units\n", + "video_file = \"amber_videos/traffic-scene.mp4\" # example video file\n", + "lens_attrs = {\"fov\": 120, \"cam_origin\": (0, 0, 0), \"skew_factor\": 0}\n", + "point_attrs = {\"p_id\": \"p1\", \"cam_id\": \"cam1\", \"x\": 0, \"y\": 0, \"z\": 0, \"time\": None, \"type\": \"pos\"}\n", + "camera_attrs = {\"ratio\": 0.5}\n", + "fps = 30\n", + "\n", + "# 1. define a world\n", + "traffic_world = empty_world(name)\n", + "\n", + "# 2. construct a camera\n", + "fov, res, cam_origin, skew_factor = (\n", + " lens_attrs[\"fov\"],\n", + " [1280, 720],\n", + " lens_attrs[\"cam_origin\"],\n", + " lens_attrs[\"skew_factor\"],\n", + ")\n", + "cam_lens = lens.PinholeLens(res, cam_origin, fov, skew_factor)\n", + "\n", + "pt_id, cam_id, x, y, z, time, pt_type = (\n", + " point_attrs[\"p_id\"],\n", + " point_attrs[\"cam_id\"],\n", + " point_attrs[\"x\"],\n", + " point_attrs[\"y\"],\n", + " point_attrs[\"z\"],\n", + " point_attrs[\"time\"],\n", + " point_attrs[\"type\"],\n", + ")\n", + "location = point.Point(pt_id, cam_id, (x, y, z), time, pt_type)\n", + "\n", + "ratio = camera_attrs[\"ratio\"]\n", + "\n", + "# ingest the camera into the world\n", + "traffic_world = traffic_world.add_camera(\n", + " cam_id=cam_id,\n", + " location=location,\n", + " ratio=ratio,\n", + " video_file=video_file,\n", + " metadata_identifier=name + \"_\" + cam_id,\n", + " lens=cam_lens,\n", + ")\n", + "\n", + "# Call execute on the world to run the detection algorithm and save the real data to the database\n", + "recognized_world = traffic_world.recognize(cam_id)\n", + "\n", + "volume = traffic_world.select_intersection_of_interest_or_use_default(cam_id=cam_id)" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "cameras are [('cam1', 0.5, 0.0, 0.0, 0.0, 369.5041722813606, 207.84609690826534, 120, 0.0)]\n", + "lens are [(0.5, 0.0, 0.0, 0.0, 120, 0.0)]\n" + ] + } + ], + "source": [ + "cams = traffic_world.get_camera()\n", + "lens = traffic_world.get_len()\n", + "# ids = traffic_world.get_id()\n", + "print(\"cameras are\", cams)\n", + "print(\"lens are\", lens)\n", + "# print(\"ids are\", ids)" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "get_traj_key SELECT sq2.itemid FROM (SELECT DISTINCT sq1.* FROM (SELECT * FROM (SELECT * FROM item_general_trajectory WHERE worldId='cd5ca1e8-dc85-45bb-9a78-fe66a587ad06') sq0 WHERE sq0.objecttype='car') sq1 CROSS JOIN cameras WHERE ST_X(cameras.origin)-getX(sq1.trajCentroids)>=-10 AND ST_X(cameras.origin)-getX(sq1.trajCentroids)<=10 AND ST_Y(cameras.origin)-getX(sq1.trajCentroids)>=-15 AND ST_Y(cameras.origin)-getX(sq1.trajCentroids)<=70 AND ST_Z(cameras.origin)-getX(sq1.trajCentroids)>=-inf AND ST_Z(cameras.origin)-getX(sq1.trajCentroids)<=inf) sq2\n" + ] + }, + { + "ename": "InFailedSqlTransaction", + "evalue": "current transaction is aborted, commands ignored until end of transaction block\n", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mInFailedSqlTransaction\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m/var/folders/jk/04mdk5z529z3kdnl2cl6ty3h0000gn/T/ipykernel_22556/1961925502.py\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[1;32m 22\u001b[0m \u001b[0;31m# and we dont care about the offset relative to the camera's z position. We also dont care about the relative heading difference.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 23\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 24\u001b[0;31m \u001b[0mfiltered_ids\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfiltered_world\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_traj_key\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 25\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"filtered_ids are\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfiltered_ids\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 26\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/Desktop/apperception/apperception/new_world.py\u001b[0m in \u001b[0;36mget_traj_key\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 168\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 169\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mget_traj_key\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 170\u001b[0;31m return derive_world(\n\u001b[0m\u001b[1;32m 171\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 172\u001b[0m \u001b[0;34m{\u001b[0m\u001b[0mType\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTRAJ\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/Desktop/apperception/apperception/new_world.py\u001b[0m in \u001b[0;36m_execute_from_root\u001b[0;34m(self, type)\u001b[0m\n\u001b[1;32m 393\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 394\u001b[0m \u001b[0;31m# print(query)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 395\u001b[0;31m \u001b[0mquery\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_execute\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mquery\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mquery\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 396\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 397\u001b[0m \u001b[0mres\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mquery\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/Desktop/apperception/apperception/new_world.py\u001b[0m in \u001b[0;36m_execute\u001b[0;34m(self, **kwargs)\u001b[0m\n\u001b[1;32m 402\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;34m\"world_id\"\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mfn_spec\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margs\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mfn_spec\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvarkw\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 403\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m\"world_id\"\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_world_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_kwargs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 404\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_kwargs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 405\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 406\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_print_lineage\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/Desktop/apperception/apperception/new_db.py\u001b[0m in \u001b[0;36mget_traj_key\u001b[0;34m(self, query)\u001b[0m\n\u001b[1;32m 193\u001b[0m \u001b[0mq\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mSnowflakeQuery\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mquery\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mselect\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"itemid\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 194\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"get_traj_key\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mq\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_sql\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 195\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcur\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexecute\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mq\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_sql\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 196\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcur\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfetchall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 197\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mInFailedSqlTransaction\u001b[0m: current transaction is aborted, commands ignored until end of transaction block\n" + ] + } + ], + "source": [ + "import time\n", + "start_time = time.time()\n", + "### Scenic Code ###\n", + "# ego = Car\n", + "# Car offset by (Range(-10, 10), Range(20, 40))\n", + "\n", + "### Apperception Query ###\n", + "filtered_world = recognized_world.filter_traj_type(\"car\")\n", + "\n", + "## OPTION 1 ###\n", + "# filtered_world = filtered_world.filter_relative_to_type(x_range=(-10, 10), y_range=(-1, 5), z_range=(-10, 0),\n", + "# type=\"camera\")\n", + "# The idea is that the user passes in a lambda function, that specifies the relationship that must be met between the queried\n", + "# object, and some object of the type passed to the function. In this case, the lambda function filters such that the offset \n", + "# is between -10 and 10 in the x direction, and between 20 and 40 in the y direction, relative to some camera.\n", + "\n", + "### OPTION 2 ###\n", + "filtered_world = filtered_world.filter_pred_relative_to_type(pred = lambda obj: (cam.x - 10) <= obj.x <= (cam.x + 10) and (cam.y - 15) <= obj.y <= (cam.y + 70))\n", + "# The idea is that filter_offset_type() takes in two arguments: the offset in terms of coordinates, a relative heading \n", + "# as well as the type of object to be offset from. In this case, we want it to be somehwere between -10 and 10 units\n", + "# offset relative to a camera's x position, somehwere between 20 and 40 units offset relative to some camera's y position, \n", + "# and we dont care about the offset relative to the camera's z position. We also dont care about the relative heading difference.\n", + "\n", + "filtered_ids = filtered_world.get_traj_key()\n", + "print(\"filtered_ids are\", filtered_ids)\n", + "\n", + "print(\"----------------------------------------------------------------------\")\n", + "print(\"Total execution time is: %s seconds\" % (time.time() - start_time))\n", + "print(\"Device Details: \\n Processor: AMD Ryzen 7 5800H \\n RAM Size: 16GB \\n Graphics Card: NVIDIA GeForce RTX 3060 Laptop\")\n", + "print(\"----------------------------------------------------------------------\")\n", + "# traffic-scene-shorter (length of 4 seconds): runtime of 81.82859063148499 seconds\n", + "# traffic-scene (length of 20 seconds): runtime of 98.58345794677734 seconds\n" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(598, 4)\n", + "(88, 4)\n", + "incorrect length: 88\n", + "(24, 4)\n", + "incorrect length: 24\n", + "output video files ./output/trafficScene_cam1car-4-a51f7975-379b-4df4-b4ce-7ba216f0b05e.mp4,./output/trafficScene_cam1car-6-a51f7975-379b-4df4-b4ce-7ba216f0b05e.mp4,./output/trafficScene_cam1car-9-a51f7975-379b-4df4-b4ce-7ba216f0b05e.mp4\n" + ] + } + ], + "source": [ + "filtered_world.get_video([cam_id], boxed=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [], + "source": [ + "# ### Scenic Code ###\n", + "# # ego = Car\n", + "# # Car\n", + "\n", + "# ### Apperception Query ###\n", + "# filtered_world = recognized_world.filter_traj_type(\"car\").interval(0, fps * 3)\n", + "# filtered_ids = filtered_world.get_traj_key()\n", + "# print(\"filtered_ids are\", filtered_ids)\n", + "\n", + "# # render tracking video\n", + "# filtered_world.get_video([cam_id])" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# ### Scenic Code ###\n", + "# # ego = Car\n", + "# # Car offset by (Range(-10, 10), Range(20, 40)), \n", + "# # \tfacing Range(-5, 5) deg\n", + "\n", + "# ### Apperception Query ###\n", + "# filtered_world = recognized_world.filter_traj_type(\"car\")\n", + "\n", + "# ## OPTION 1 ###\n", + "# filtered_world = filtered_world.filter_relative_to_type(relative=lambda obj, camera: -10 <= (camera.x - obj.x) <= 10 \\\n", + "# and 20 <= (camera.y - obj.y) <= 40,\n", + "# type=\"camera\")\n", + "\n", + "# ### OPTION 2 ###\n", + "# filtered_world = filtered_world.filter_relative_to_type(offset=((-10, 10), (20, 40), None), heading=None, type=\"camera\")\n", + "\n", + "# filtered_world = filtered_world.filter_heading(-5, 5)\n", + "# # Filters for objects that have heading between -5 and 5 degrees\n", + "\n", + "# filtered_ids = filtered_world.get_traj_key()\n", + "# print(\"filtered_ids are\", filtered_ids)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# ### Scenic Code ###\n", + "# # ego = Car\n", + "# # Car offset by (Range(-10, 10), Range(20, 40)), \n", + "# # \tfacing Range(-5, 5) deg relative to ego\n", + "\n", + "# ### Apperception Query ###\n", + "# filtered_world = recognized_world.filter_traj_type(\"car\")\n", + "\n", + "# ## OPTION 1 ###\n", + "# filtered_world = filtered_world.filter_relative_to_type(relative=lambda obj, camera: -10 <= (camera.x - obj.x) <= 10 \\\n", + "# and 20 <= (camera.y - obj.y) <= 40 \\\n", + "# and -5 <= (camera.heading - obj.heading) <= 5,\n", + "# type=\"camera\")\n", + "# # Now filtering for a relative heading between -5 and 5 degrees\n", + "\n", + "# ### OPTION 2 ###\n", + "# filtered_world = filtered_world.filter_relative_to_type(offset=((-10, 10), (20, 40), None), heading=(-5, 5), type=\"camera\")\n", + "# # Now filtering for a relative heading between -5 and 5 degrees\n", + "\n", + "# filtered_ids = filtered_world.get_traj_key()\n", + "# print(\"filtered_ids are\", filtered_ids)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# ### Scenic Code ###\n", + "# # ego = Car\n", + "# # Car left of ego by 0.25 \n", + "\n", + "# ### Apperception Query ###\n", + "# filtered_world = recognized_world.filter_traj_type(\"car\")\n", + "\n", + "# ## OPTION 1 ###\n", + "# def left_of(obj, camera):\n", + "# expec_x = obj.x + 0.25 * np.cos(camera.heading)\n", + "# expec_y = obj.y - 0.25 * np.sin(camera.heading)\n", + "# # Should also allow some sort of variation, to account for noise (and since exact equality is unlikley)\n", + "# return (expec_x == camera.x) and (expec_y == camera.y)\n", + "\n", + "# filtered_world = filtered_world.filter_relative_to_type(relative=left_of, type=\"camera\")\n", + "# # Now filtering such that the car is left of ego by 0.25 units\n", + "\n", + "# ### OPTION 2 ##\n", + "# # Not possible\n", + "\n", + "# filtered_ids = filtered_world.get_traj_key()\n", + "# print(\"filtered_ids are\", filtered_ids)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# ### Scenic Code ###\n", + "# # ego = Car\n", + "# # badAngle = Range(10, 20) deg\n", + "# # Car left of ego by 0.25,\n", + "# # \tfacing badAngle relative to ego\n", + "\n", + "# ### Apperception Query ###\n", + "# filtered_world = recognized_world.filter_traj_type(\"car\")\n", + "\n", + "# ## OPTION 1 ###\n", + "# def filter(obj, camera):\n", + "# expec_x = obj.x + 0.25 * np.cos(camera.heading)\n", + "# expec_y = obj.y - 0.25 * np.sin(camera.heading)\n", + "# # Should also allow some sort of variation, to account for noise (and since exact equality is unlikley)\n", + "# return (expec_x == camera.x) and (expec_y == camera.y) and 10 <= (camera.heading - obj.heading) <= 20\n", + "\n", + "# filtered_world = filtered_world.filter_relative_to_type(relative=filter, type=\"camera\")\n", + "# # Now filtering such that the car is left of ego by 0.25 units\n", + "\n", + "# ### OPTION 2 ##\n", + "# # Not possible\n", + "\n", + "# filtered_ids = filtered_world.get_traj_key()\n", + "# print(\"filtered_ids are\", filtered_ids)" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": {}, + "outputs": [], + "source": [ + "# def roadDirection(x, y, z):\n", + "# # TODO: Implement\n", + "# # Returns the direction (in 360 degree angle form) of the road at the coordinates (x, y, z)\n", + "# # If their is no such road, returns a value of None\n", + "# return None" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "#### FURTHER QUERIES WILL USE THE OPTION 1 LISTED ABOVE ####" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Scenic Code ###\n", + "# weather = Uniform(\"sunny\", \"rainy\", \"thunder\")\n", + "# time = Range(10, 12)\n", + "#\n", + "# ego = car on road\n", + "# otherCar = Car ahead of ego by Range(4, 19)\n", + "# require not (otherCar in intersection)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Scenic Code ###\n", + "# spot = OrientedPoint on curb\n", + "# ego = Car at (spot offset by (Range(2,4), Range(5,10)))\n", + "# sideCar = Car left of spot by Range(1,3)" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [], + "source": [ + "### Scenic Code ###\n", + "# def placeObjs(car, numCars):\n", + "# for i in range(numCars):\n", + "# car = Car ahead of car by Range(4, 5)\n", + "# leftCar = Car left of car by Normal(2, 0.1), facing roadDirection\n", + "# rightCar = Car right of car by Normal(3, 0.1), facing Range(0, 10) deg relative to ego.heading\n", + "# return leftCar, rightCar\n", + "\n", + "# spawn_point = 207.26 @ 8.72\n", + "# ego = Car at spawn_point, with visible_distance 200\n", + "\n", + "# leftCar, rightCar = placeObjs(ego, 2)\n", + "# require (distance to leftCar) < 200\n", + "# require (distance to rightCar) < 200\n" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [], + "source": [ + "### Scenic Code ###\n", + "# def placeObjs(numPeds):\n", + "# for i in range(numPeds):\n", + "# Pedestrian offset by Range(-5, 5) @ Range(0, 200),\n", + "# facing Range(-120, 120) deg relative to ego.heading\n", + "\n", + "# spawn_point = 207.26 @ 8.72\n", + "# ego = Car at spawn_point,\n", + "# with visibleDistance 200\n", + "\n", + "# placeObjs(3)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Scenic Code ###\n", + "# ego = Car on drivableRoad,\n", + "# facing Range(-15, 15) deg relative to roadDirection,\n", + "# with visibleDistance 50, \n", + "# with viewAngle 135 deg\n", + "# ped = Pedestrian on roadsOrIntersections,\n", + "# with regionContainedIn roadRegion,\n", + "# facing Range(-180, 180) deg\n", + "\n", + "# require abs(relative heading of ped from ego) > 70 deg" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [], + "source": [ + "### Scenic Code ###\n", + "# offset = Uniform(-1, 1) * Range(90, 180) deg\n", + "\n", + "# ego = Car on drivableRoad,\n", + "# facing offset relative to roadDirection,\n", + "# with visibleDistance 50,\n", + "# with viewAngle 135 deg\n", + "\n", + "# otherCar = Car on visible road,\n", + "# facing Range(-15, 15) deg relative to roadDirection\n", + "\n", + "# require (distance from ego to otherCar) < 10" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [], + "source": [ + "### Scenic Code ###\n", + "# ego = Car on drivableRoad,\n", + "# facing Range(-15, 15) deg relative to roadDirection,\n", + "# with visibleDistance 50,\n", + "# with viewAngle 135 deg\n", + "\n", + "# other1 = Car on intersection,\n", + "# facing -1 * Range(50, 135) deg relative to ego.heading\n", + "\n", + "# other2 = Car on intersection,\n", + "# facing -1 * Range(50, 135) deg relative to ego.heading\n", + "\n", + "# require abs(relative heading of other1 from other2) > 100 deg\n", + "# require (distance from ego to intersectionRegion) < 10" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [], + "source": [ + "### Scenic Code ###\n", + "# ego = Car on drivableRoad,\n", + "# facing Range(-15, 15) deg relative to roadDirection,\n", + "# with visibleDistance 50,\n", + "# with viewAngle 135 deg\n", + "\n", + "# point1 = OrientedPoint ahead of ego by Range(0, 40)\n", + "# Car at (point1 offset by Range(-1, 1) & 0),\n", + "# facing Range(-15, 15) deg relative to roadDirection\n", + "\n", + "# oppositeCar = Car offset by (Range(-10, -1), Range(0, 50)),\n", + "# facing Range(140, 180) deg relative to ego.heading\n", + "\n", + "# point2 = OrientedPoint ahead of oppositeCar by Range(0, 40)\n", + "# Car at (point2 offset by Range(-1, 1) @ 0),\n", + "# facing Range(-15, 15) deg relative to roadDirection" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": {}, + "outputs": [], + "source": [ + "### Scenic Code ###\n", + "# lanesWithRightLane = filter(lambda i: i._laneToRight, network.laneSections)\n", + "# egoLane = Uniform(*lanesWithRightLane)\n", + "\n", + "# ego = Car on egoLane,\n", + "# facing Range(-15, 15) deg relative to roadDirection\n", + "# cutInCar = Car offset by Range(0, 4) @ Range(0, 5),\n", + "# facing -1*Range(15, 30) deg relative to roadDirection" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [], + "source": [ + "# I think there are 3 main things that need to now be implemented in Apperception to allow incorparation with Scenic:\n", + "# 1. A way to have the arbitrary filters that were possible in the old API (the predicate lambda functions that could be passed)\n", + "# 2. A way to have filters with regardes to other objects. For example, I could say I want \"cars that are to the left of a bus by 0.25m\" or such. I would assume this could also be implemented as a lambda function filter (I have included an example fo this in the scenic_equivelants notebook).\n", + "# 3. Some way to not only recognize what the type of an object is, but recognize the type of point it is on. For example, recognizing that the Car is on a road, or that the Car is in an intersection (this is something that is done quite a lot in Scenic).\n", + "# - For this, we might not have to incorporate it into apperception, and can make it the users responsibility (and they can create their own filters that do this), but I am not too sure" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "interpreter": { + "hash": "36b9f45d2c0c5940d48526f9dac9a46c8afda5d718c8f108cd3f22cd85be16c2" + }, + "kernelspec": { + "display_name": "Python 3.8.8 ('base')", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.12" + }, + "orig_nbformat": 4 + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/scenic_equivelants.py b/scenic_equivelants.py new file mode 100644 index 00000000..db7fb548 --- /dev/null +++ b/scenic_equivelants.py @@ -0,0 +1,367 @@ +# %% +# Deleting the .apperception_cache if it exists, as to avoid DB conflict errors +import os +import shutil + +dirpath = os.path.join('.apperception_cache') +if os.path.exists(dirpath) and os.path.isdir(dirpath): + shutil.rmtree(dirpath) + +dirpath = os.path.join('output') +if os.path.exists(dirpath) and os.path.isdir(dirpath): + shutil.rmtree(dirpath) +os.mkdir(dirpath) + +# This piece of code is unsafe, and should not be run if not needed. +# It serves for test purposes when one recieves a "dead kernel" error. +os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE" + +# %% +import sys +sys.path.append(os.path.join(os.getcwd(),"apperception")) + +### IMPORTS +import lens +import point +from new_world import empty_world + +# Let's define some attribute for constructing the world first +name = "trafficScene" # world name +units = "metrics" # world units +video_file = "amber_videos/traffic-scene.mp4" # example video file +lens_attrs = {"fov": 120, "cam_origin": (0, 0, 0), "skew_factor": 0} +point_attrs = {"p_id": "p1", "cam_id": "cam1", "x": 0, "y": 0, "z": 0, "time": None, "type": "pos"} +camera_attrs = {"ratio": 0.5} +fps = 30 + +# 1. define a world +traffic_world = empty_world(name) + +# 2. construct a camera +fov, res, cam_origin, skew_factor = ( + lens_attrs["fov"], + [1280, 720], + lens_attrs["cam_origin"], + lens_attrs["skew_factor"], +) +cam_lens = lens.PinholeLens(res, cam_origin, fov, skew_factor) + +pt_id, cam_id, x, y, z, time, pt_type = ( + point_attrs["p_id"], + point_attrs["cam_id"], + point_attrs["x"], + point_attrs["y"], + point_attrs["z"], + point_attrs["time"], + point_attrs["type"], +) +location = point.Point(pt_id, cam_id, (x, y, z), time, pt_type) + +ratio = camera_attrs["ratio"] + +# ingest the camera into the world +traffic_world = traffic_world.add_camera( + cam_id=cam_id, + location=location, + ratio=ratio, + video_file=video_file, + metadata_identifier=name + "_" + cam_id, + lens=cam_lens, +) + +# Call execute on the world to run the detection algorithm and save the real data to the database +recognized_world = traffic_world.recognize(cam_id) + +volume = traffic_world.select_intersection_of_interest_or_use_default(cam_id=cam_id) + +# %% +cams = traffic_world.get_camera() +lens = traffic_world.get_len() +# ids = traffic_world.get_id() +print("cameras are", cams) +print("lens are", lens) +# print("ids are", ids) + +# %% +import time +start_time = time.time() +### Scenic Code ### +# ego = Car +# Car offset by (Range(-10, 10), Range(20, 40)) + +### Apperception Query ### +filtered_world = recognized_world.filter_traj_type("car") + +## OPTION 1 ### +# filtered_world = filtered_world.filter_relative_to_type(x_range=(-10, 10), y_range=(-1, 5), z_range=(-10, 0), +# type="camera") +# The idea is that the user passes in a lambda function, that specifies the relationship that must be met between the queried +# object, and some object of the type passed to the function. In this case, the lambda function filters such that the offset +# is between -10 and 10 in the x direction, and between 20 and 40 in the y direction, relative to some camera. + +### OPTION 2 ### +filtered_world = filtered_world.filter_pred_relative_to_type(pred = lambda obj: (cam.x - 10) <= obj.x <= (cam.x + 10) and (cam.y - 15) <= obj.y <= (cam.y + 70)) +# The idea is that filter_offset_type() takes in two arguments: the offset in terms of coordinates, a relative heading +# as well as the type of object to be offset from. In this case, we want it to be somehwere between -10 and 10 units +# offset relative to a camera's x position, somehwere between 20 and 40 units offset relative to some camera's y position, +# and we dont care about the offset relative to the camera's z position. We also dont care about the relative heading difference. + +filtered_ids = filtered_world.get_traj_key() +print("filtered_ids are", filtered_ids) + +print("----------------------------------------------------------------------") +print("Total execution time is: %s seconds" % (time.time() - start_time)) +print("Device Details: \n Processor: AMD Ryzen 7 5800H \n RAM Size: 16GB \n Graphics Card: NVIDIA GeForce RTX 3060 Laptop") +print("----------------------------------------------------------------------") +# traffic-scene-shorter (length of 4 seconds): runtime of 81.82859063148499 seconds +# traffic-scene (length of 20 seconds): runtime of 98.58345794677734 seconds + + +# %% +filtered_world.get_video([cam_id], boxed=True) + +# %% +# ### Scenic Code ### +# # ego = Car +# # Car + +# ### Apperception Query ### +# filtered_world = recognized_world.filter_traj_type("car").interval(0, fps * 3) +# filtered_ids = filtered_world.get_traj_key() +# print("filtered_ids are", filtered_ids) + +# # render tracking video +# filtered_world.get_video([cam_id]) + +# %% +# ### Scenic Code ### +# # ego = Car +# # Car offset by (Range(-10, 10), Range(20, 40)), +# # facing Range(-5, 5) deg + +# ### Apperception Query ### +# filtered_world = recognized_world.filter_traj_type("car") + +# ## OPTION 1 ### +# filtered_world = filtered_world.filter_relative_to_type(relative=lambda obj, camera: -10 <= (camera.x - obj.x) <= 10 \ +# and 20 <= (camera.y - obj.y) <= 40, +# type="camera") + +# ### OPTION 2 ### +# filtered_world = filtered_world.filter_relative_to_type(offset=((-10, 10), (20, 40), None), heading=None, type="camera") + +# filtered_world = filtered_world.filter_heading(-5, 5) +# # Filters for objects that have heading between -5 and 5 degrees + +# filtered_ids = filtered_world.get_traj_key() +# print("filtered_ids are", filtered_ids) + +# %% +# ### Scenic Code ### +# # ego = Car +# # Car offset by (Range(-10, 10), Range(20, 40)), +# # facing Range(-5, 5) deg relative to ego + +# ### Apperception Query ### +# filtered_world = recognized_world.filter_traj_type("car") + +# ## OPTION 1 ### +# filtered_world = filtered_world.filter_relative_to_type(relative=lambda obj, camera: -10 <= (camera.x - obj.x) <= 10 \ +# and 20 <= (camera.y - obj.y) <= 40 \ +# and -5 <= (camera.heading - obj.heading) <= 5, +# type="camera") +# # Now filtering for a relative heading between -5 and 5 degrees + +# ### OPTION 2 ### +# filtered_world = filtered_world.filter_relative_to_type(offset=((-10, 10), (20, 40), None), heading=(-5, 5), type="camera") +# # Now filtering for a relative heading between -5 and 5 degrees + +# filtered_ids = filtered_world.get_traj_key() +# print("filtered_ids are", filtered_ids) + +# %% +# ### Scenic Code ### +# # ego = Car +# # Car left of ego by 0.25 + +# ### Apperception Query ### +# filtered_world = recognized_world.filter_traj_type("car") + +# ## OPTION 1 ### +# def left_of(obj, camera): +# expec_x = obj.x + 0.25 * np.cos(camera.heading) +# expec_y = obj.y - 0.25 * np.sin(camera.heading) +# # Should also allow some sort of variation, to account for noise (and since exact equality is unlikley) +# return (expec_x == camera.x) and (expec_y == camera.y) + +# filtered_world = filtered_world.filter_relative_to_type(relative=left_of, type="camera") +# # Now filtering such that the car is left of ego by 0.25 units + +# ### OPTION 2 ## +# # Not possible + +# filtered_ids = filtered_world.get_traj_key() +# print("filtered_ids are", filtered_ids) + +# %% +# ### Scenic Code ### +# # ego = Car +# # badAngle = Range(10, 20) deg +# # Car left of ego by 0.25, +# # facing badAngle relative to ego + +# ### Apperception Query ### +# filtered_world = recognized_world.filter_traj_type("car") + +# ## OPTION 1 ### +# def filter(obj, camera): +# expec_x = obj.x + 0.25 * np.cos(camera.heading) +# expec_y = obj.y - 0.25 * np.sin(camera.heading) +# # Should also allow some sort of variation, to account for noise (and since exact equality is unlikley) +# return (expec_x == camera.x) and (expec_y == camera.y) and 10 <= (camera.heading - obj.heading) <= 20 + +# filtered_world = filtered_world.filter_relative_to_type(relative=filter, type="camera") +# # Now filtering such that the car is left of ego by 0.25 units + +# ### OPTION 2 ## +# # Not possible + +# filtered_ids = filtered_world.get_traj_key() +# print("filtered_ids are", filtered_ids) + +# %% +# def roadDirection(x, y, z): +# # TODO: Implement +# # Returns the direction (in 360 degree angle form) of the road at the coordinates (x, y, z) +# # If their is no such road, returns a value of None +# return None + +# %% +#### FURTHER QUERIES WILL USE THE OPTION 1 LISTED ABOVE #### + +# %% +### Scenic Code ### +# weather = Uniform("sunny", "rainy", "thunder") +# time = Range(10, 12) +# +# ego = car on road +# otherCar = Car ahead of ego by Range(4, 19) +# require not (otherCar in intersection) + +# %% +### Scenic Code ### +# spot = OrientedPoint on curb +# ego = Car at (spot offset by (Range(2,4), Range(5,10))) +# sideCar = Car left of spot by Range(1,3) + +# %% +### Scenic Code ### +# def placeObjs(car, numCars): +# for i in range(numCars): +# car = Car ahead of car by Range(4, 5) +# leftCar = Car left of car by Normal(2, 0.1), facing roadDirection +# rightCar = Car right of car by Normal(3, 0.1), facing Range(0, 10) deg relative to ego.heading +# return leftCar, rightCar + +# spawn_point = 207.26 @ 8.72 +# ego = Car at spawn_point, with visible_distance 200 + +# leftCar, rightCar = placeObjs(ego, 2) +# require (distance to leftCar) < 200 +# require (distance to rightCar) < 200 + + +# %% +### Scenic Code ### +# def placeObjs(numPeds): +# for i in range(numPeds): +# Pedestrian offset by Range(-5, 5) @ Range(0, 200), +# facing Range(-120, 120) deg relative to ego.heading + +# spawn_point = 207.26 @ 8.72 +# ego = Car at spawn_point, +# with visibleDistance 200 + +# placeObjs(3) + +# %% +### Scenic Code ### +# ego = Car on drivableRoad, +# facing Range(-15, 15) deg relative to roadDirection, +# with visibleDistance 50, +# with viewAngle 135 deg +# ped = Pedestrian on roadsOrIntersections, +# with regionContainedIn roadRegion, +# facing Range(-180, 180) deg + +# require abs(relative heading of ped from ego) > 70 deg + +# %% +### Scenic Code ### +# offset = Uniform(-1, 1) * Range(90, 180) deg + +# ego = Car on drivableRoad, +# facing offset relative to roadDirection, +# with visibleDistance 50, +# with viewAngle 135 deg + +# otherCar = Car on visible road, +# facing Range(-15, 15) deg relative to roadDirection + +# require (distance from ego to otherCar) < 10 + +# %% +### Scenic Code ### +# ego = Car on drivableRoad, +# facing Range(-15, 15) deg relative to roadDirection, +# with visibleDistance 50, +# with viewAngle 135 deg + +# other1 = Car on intersection, +# facing -1 * Range(50, 135) deg relative to ego.heading + +# other2 = Car on intersection, +# facing -1 * Range(50, 135) deg relative to ego.heading + +# require abs(relative heading of other1 from other2) > 100 deg +# require (distance from ego to intersectionRegion) < 10 + +# %% +### Scenic Code ### +# ego = Car on drivableRoad, +# facing Range(-15, 15) deg relative to roadDirection, +# with visibleDistance 50, +# with viewAngle 135 deg + +# point1 = OrientedPoint ahead of ego by Range(0, 40) +# Car at (point1 offset by Range(-1, 1) & 0), +# facing Range(-15, 15) deg relative to roadDirection + +# oppositeCar = Car offset by (Range(-10, -1), Range(0, 50)), +# facing Range(140, 180) deg relative to ego.heading + +# point2 = OrientedPoint ahead of oppositeCar by Range(0, 40) +# Car at (point2 offset by Range(-1, 1) @ 0), +# facing Range(-15, 15) deg relative to roadDirection + +# %% +### Scenic Code ### +# lanesWithRightLane = filter(lambda i: i._laneToRight, network.laneSections) +# egoLane = Uniform(*lanesWithRightLane) + +# ego = Car on egoLane, +# facing Range(-15, 15) deg relative to roadDirection +# cutInCar = Car offset by Range(0, 4) @ Range(0, 5), +# facing -1*Range(15, 30) deg relative to roadDirection + +# %% +# I think there are 3 main things that need to now be implemented in Apperception to allow incorparation with Scenic: +# 1. A way to have the arbitrary filters that were possible in the old API (the predicate lambda functions that could be passed) +# 2. A way to have filters with regardes to other objects. For example, I could say I want "cars that are to the left of a bus by 0.25m" or such. I would assume this could also be implemented as a lambda function filter (I have included an example fo this in the scenic_equivelants notebook). +# 3. Some way to not only recognize what the type of an object is, but recognize the type of point it is on. For example, recognizing that the Car is on a road, or that the Car is in an intersection (this is something that is done quite a lot in Scenic). +# - For this, we might not have to incorporate it into apperception, and can make it the users responsibility (and they can create their own filters that do this), but I am not too sure + +# %% + + + diff --git a/setup.sh b/setup.sh index 0519ae1a..834c8ae1 100755 --- a/setup.sh +++ b/setup.sh @@ -16,7 +16,11 @@ mv checkpoints/ ./yolov4-deepsort cp ./configs/yolov4-config.py ./yolov4-deepsort/core/config.py # setup YoloV5 -git clone --recurse-submodules git@github.com:mikel-brostrom/Yolov5_DeepSort_Pytorch.git yolov5-deepsort +yolov5_dir="yolov5-deepsort" +git clone --recurse-submodules git@github.com:mikel-brostrom/Yolov5_DeepSort_Pytorch.git "${yolov5_dir}" +pushd "${yolov5_dir}" +git checkout 8aec0b5 +popd cp ./configs/yolov5-deepsort-config.yaml ./yolov5-deepsort/deep_sort_pytorch/configs/deep_sort.yaml pushd yolov5-deepsort python3 -m pip install -r requirements.txt From 88cd5fcc92db44d90f09378380f47ccac9538a5a Mon Sep 17 00:00:00 2001 From: "Chanwut (Mick) Kittivorawong" Date: Tue, 8 Mar 2022 06:53:47 -0800 Subject: [PATCH 2/4] move all the benchmarks --- {apperception => benchmarks}/apperception_benchmark.py | 0 .../apperception_example.ipynb | 0 .../new_apperception_benchmark.ipynb | 0 {apperception => benchmarks}/new_apperception_benchmark.py | 0 {apperception => benchmarks}/scenic_benchmark.py | 0 .../scenic_equivelants.ipynb | 0 scenic_test.ipynb => benchmarks/scenic_test.ipynb | 0 requirements.txt | 4 +++- 8 files changed, 3 insertions(+), 1 deletion(-) rename {apperception => benchmarks}/apperception_benchmark.py (100%) rename apperception_example.ipynb => benchmarks/apperception_example.ipynb (100%) rename new_apperception_benchmark.ipynb => benchmarks/new_apperception_benchmark.ipynb (100%) rename {apperception => benchmarks}/new_apperception_benchmark.py (100%) rename {apperception => benchmarks}/scenic_benchmark.py (100%) rename scenic_equivelants.ipynb => benchmarks/scenic_equivelants.ipynb (100%) rename scenic_test.ipynb => benchmarks/scenic_test.ipynb (100%) diff --git a/apperception/apperception_benchmark.py b/benchmarks/apperception_benchmark.py similarity index 100% rename from apperception/apperception_benchmark.py rename to benchmarks/apperception_benchmark.py diff --git a/apperception_example.ipynb b/benchmarks/apperception_example.ipynb similarity index 100% rename from apperception_example.ipynb rename to benchmarks/apperception_example.ipynb diff --git a/new_apperception_benchmark.ipynb b/benchmarks/new_apperception_benchmark.ipynb similarity index 100% rename from new_apperception_benchmark.ipynb rename to benchmarks/new_apperception_benchmark.ipynb diff --git a/apperception/new_apperception_benchmark.py b/benchmarks/new_apperception_benchmark.py similarity index 100% rename from apperception/new_apperception_benchmark.py rename to benchmarks/new_apperception_benchmark.py diff --git a/apperception/scenic_benchmark.py b/benchmarks/scenic_benchmark.py similarity index 100% rename from apperception/scenic_benchmark.py rename to benchmarks/scenic_benchmark.py diff --git a/scenic_equivelants.ipynb b/benchmarks/scenic_equivelants.ipynb similarity index 100% rename from scenic_equivelants.ipynb rename to benchmarks/scenic_equivelants.ipynb diff --git a/scenic_test.ipynb b/benchmarks/scenic_test.ipynb similarity index 100% rename from scenic_test.ipynb rename to benchmarks/scenic_test.ipynb diff --git a/requirements.txt b/requirements.txt index 55ee9da9..fb33b64f 100644 --- a/requirements.txt +++ b/requirements.txt @@ -14,4 +14,6 @@ matplotlib pillow jupyter torch -numpy \ No newline at end of file +numpy +pyquaternion +PyQt5 \ No newline at end of file From c2132f681ed936ea392cb2d7ebe7baa2d810e3e8 Mon Sep 17 00:00:00 2001 From: "Chanwut (Mick) Kittivorawong" Date: Sun, 13 Mar 2022 01:37:26 -0800 Subject: [PATCH 3/4] fix rebase error --- apperception/camera.py | 14 + apperception/camera_config.py | 40 +++ apperception/new_db.py | 261 +++++++++--------- apperception/new_util.py | 217 +++++++++++++++ apperception/new_world.py | 82 +++--- .../object_tracker_yolov4_deepsort.py | 2 +- .../object_tracker_yolov5_deepsort.py | 2 +- apperception/scenic_util.py | 7 +- apperception/tracked_object.py | 33 ++- apperception/video_context.py | 4 +- apperception/video_context_executor.py | 2 +- apperception/video_util.py | 2 +- benchmarks/new-api-scenic.ipynb | 164 +++++++++++ 13 files changed, 638 insertions(+), 192 deletions(-) create mode 100644 apperception/camera.py create mode 100644 apperception/camera_config.py create mode 100644 benchmarks/new-api-scenic.ipynb diff --git a/apperception/camera.py b/apperception/camera.py new file mode 100644 index 00000000..2acdef73 --- /dev/null +++ b/apperception/camera.py @@ -0,0 +1,14 @@ +from dataclasses import dataclass +from typing import List, Optional +import uuid + +from camera_config import CameraConfig + +@dataclass +class Camera: + id: str + configs: List[CameraConfig] + + def __init__(self, config: List[CameraConfig], id: Optional[str] = None): + self.id = str(uuid.uuid4()) if id is None else id + self.configs = config diff --git a/apperception/camera_config.py b/apperception/camera_config.py new file mode 100644 index 00000000..e6bed4d6 --- /dev/null +++ b/apperception/camera_config.py @@ -0,0 +1,40 @@ +from dataclasses import dataclass +from typing import List, Tuple + +Float3 = Tuple[float, float, float] +Float4 = Tuple[float, float, float, float] + +@dataclass(frozen=True) +class CameraConfig: + frame_id: str + frame_num: int + filename: str + camera_translation: List[float] # float[3] + camera_rotation: List[float] # float[4] + camera_intrinsic: List[List[float]] # float[3][3] + ego_translation: List[float] # float[3] + ego_rotation: List[float] # float[4] + timestamp: str + + +def fetch_camera_config(scene_name: str, sample_data): + all_frames = sample_data[ + (sample_data["scene_name"] == scene_name) + & (sample_data["filename"].str.contains("/CAM_FRONT/", regex=False)) + ].sort_values(by="frame_order") + + return [ + CameraConfig( + frame_id=frame.sample_token, + frame_num=frame.frame_order, + filename=frame.filename, + camera_translation=frame.camera_translation, + camera_rotation=frame.camera_rotation, + camera_intrinsic=frame.camera_intrinsic, + ego_translation=frame.ego_translation, + ego_rotation=frame.ego_rotation, + timestamp=frame.timestamp, + ) + for frame + in all_frames.itertuples(index=False) + ] \ No newline at end of file diff --git a/apperception/new_db.py b/apperception/new_db.py index 4fbffa30..e3e87623 100644 --- a/apperception/new_db.py +++ b/apperception/new_db.py @@ -4,39 +4,34 @@ import psycopg2 from bounding_box import BoundingBox from lens import PinholeLens -from new_util import create_camera, get_video, video_fetch_reformat +from new_util import create_camera, get_video, video_fetch_reformat, add_recognized_objs, recognize from pypika import Column, CustomFunction, Table # https://github.com/kayak/pypika/issues/553 # workaround. because the normal Query will fail due to mobility db from pypika.dialects import Query, SnowflakeQuery -from video_context import Camera -from video_util import add_recognized_objs, get_video_dimension, recognize +from camera import Camera -CAMERA_TABLE = "cameras" -TRAJ_TABLE = "item_general_trajectory" -BBOX_TABLE = "general_bbox" +CAMERA_TABLE = "Cameras" +TRAJ_TABLE = "Item_General_Trajectory" +BBOX_TABLE = "General_Bbox" class Database: - def __init__(self, reset: bool = True): + def __init__(self): # should setup a postgres in docker first self.con = psycopg2.connect( dbname="mobilitydb", user="docker", host="localhost", port="25432", password="docker" ) self.cur = self.con.cursor() - if reset: - # create camera table - self._create_camera_table() - - # create bbox table - self._create_general_bbox_table() - - # create traj table - self._create_item_general_trajectory_table() - # The start time of the database access object self.start_time = datetime.datetime(2021, 6, 8, 7, 10, 28) + + def reset(self): + self._create_camera_table() + self._create_item_general_trajectory_table() + self._create_general_bbox_table() + self._create_index() def _create_camera_table(self): # drop old @@ -45,70 +40,136 @@ def _create_camera_table(self): # create new q2 = SnowflakeQuery.create_table(CAMERA_TABLE).columns( Column("cameraId", "TEXT"), - Column("worldId", "TEXT"), - Column("ratio", "real"), - Column("origin", "geometry"), - Column("focalpoints", "geometry"), - Column("fov", "INTEGER"), - Column("skev_factor", "real"), - Column("width", "integer"), - Column("height", "integer"), + Column("frameId", "TEXT"), + Column("frameNum", "Int"), + Column("fileName", "TEXT"), + Column("cameraTranslation", "geometry"), + Column("cameraRotation", "real[4]"), + Column("cameraIntrinsic", "real[3][3]"), + Column("egoTranslation", "geometry"), + Column("egoRotation", "real[4]"), + Column("timestamp", "TEXT"), ) self.cur.execute(q1.get_sql()) self.cur.execute(q2.get_sql()) self.con.commit() - def insert_cam(self, world_id: str, camera_node: Camera): - cam = Table(CAMERA_TABLE) - cam_id = camera_node.cam_id - cam_ratio = camera_node.ratio - lens = camera_node.lens - - if not isinstance(lens, PinholeLens): - raise Exception("Only accept a camera with PinholeLens") - - focal_x = str(lens.focal_x) - focal_y = str(lens.focal_y) - cam_x, cam_y, cam_z = ( - str(lens.cam_origin[0]), - str(lens.cam_origin[1]), - str(lens.cam_origin[2]), - ) - width, height = get_video_dimension(camera_node.video_file) - - q = SnowflakeQuery.into(cam).insert( - cam_id, - world_id, - cam_ratio, - f"POINT Z ({cam_x} {cam_y} {cam_z})", - f"POINT({focal_x} {focal_y})", - lens.fov, - lens.alpha, - width, - height, - ) - # print(q) - self.cur.execute(q.get_sql()) + def _create_general_bbox_table(self): + # drop old + q1 = SnowflakeQuery.drop_table(BBOX_TABLE).if_exists() + + # create new + q2 = """ + CREATE TABLE General_Bbox( + itemId TEXT, + cameraId TEXT, + trajBbox stbox, + FOREIGN KEY(itemId) + REFERENCES Item_General_Trajectory(itemId) + ); + """ + + self.cur.execute(q1.get_sql()) + self.cur.execute(q2) + self.con.commit() + + def _create_item_general_trajectory_table(self): + # drop old + q1 = 'DROP TABLE IF EXISTS Item_General_Trajectory CASCADE;' + + # create new + q2 = """ + CREATE TABLE Item_General_Trajectory( + itemId TEXT, + cameraId TEXT, + objectType TEXT, + frameId TEXT, + color TEXT, + trajCentroids tgeompoint, + largestBbox stbox, + PRIMARY KEY (itemId) + ); + """ + + self.cur.execute(q1) + self.cur.execute(q2) self.con.commit() + + def _create_index(self): + self.cur.execute(""" + CREATE INDEX IF NOT EXISTS traj_idx + ON Item_General_Trajectory + USING GiST(trajCentroids); + """) + self.cur.execute(""" + CREATE INDEX IF NOT EXISTS item_idx + ON General_Bbox(itemId); + """) + self.cur.execute(""" + CREATE INDEX IF NOT EXISTS traj_bbox_idx + ON General_Bbox + USING GiST(trajBbox); + """) + self.con.commit() + + + def insert_cam(self, camera: Camera): + values = [ + f"""( + '{camera.id}', + '{config.frame_id}', + {config.frame_num}, + '{config.filename}', + 'POINT Z ({' '.join(map(str, config.camera_translation))})', + ARRAY{config.camera_rotation}, + ARRAY{config.camera_intrinsic}, + 'POINT Z ({' '.join(map(str, config.ego_translation))})', + ARRAY{config.ego_rotation}, + '{config.timestamp}' + )""" + for config + in camera.configs + ] + + self.cur.execute( + f""" + INSERT INTO Cameras ( + cameraId, + frameId, + frameNum, + fileName, + cameraTranslation, + cameraRotation, + cameraIntrinsic, + egoTranslation, + egoRotation, + timestamp + ) + VALUES {','.join(values)}; + """ + ) + + print("New camera inserted successfully.........") + self.conn.commit() - def retrieve_cam(self, query: Query = None, world_id: str = ""): + def retrieve_cam(self, query: Query = None, camera_id: str = ""): """ Called when executing update commands (add_camera, add_objs ...etc) """ return ( - query + self._select_cam_with_world_id(world_id) + query + self._select_cam_with_camera_id(camera_id) if query - else self._select_cam_with_world_id(world_id) + else self._select_cam_with_camera_id(camera_id) ) # UNION - def _select_cam_with_world_id(self, world_id: str): + def _select_cam_with_camera_id(self, camera_id: str): """ Select cams with certain world id """ cam = Table(CAMERA_TABLE) - q = SnowflakeQuery.from_(cam).select("*").where(cam.worldId == world_id) + q = SnowflakeQuery.from_(cam).select("*").where(cam.id == camera_id) return q def filter_cam(self, query: Query, condition: str): @@ -147,31 +208,18 @@ def get_len(self, query: Query): self.cur.execute(q) return self.cur.fetchall() - def _create_general_bbox_table(self): - # already created in create_or_insert_general_trajectory - self.cur.execute("DROP TABLE IF EXISTS General_Bbox;") - self.con.commit() - - def _create_item_general_trajectory_table(self): - # already created in create_or_insert_general_trajectory - self.cur.execute("DROP TABLE IF EXISTS Item_General_Trajectory;") - self.con.commit() - - def insert_bbox_traj(self, world_id: str, camera_node: Camera, recognition_area: BoundingBox): - video_file, algo, lens = camera_node.video_file, "Yolo", camera_node.lens - tracking_results = recognize( - video_file=video_file, recog_algo=algo, recognition_area=recognition_area - ) - add_recognized_objs(self.con, lens, tracking_results, self.start_time, world_id) + def insert_bbox_traj(self, camera: Camera, annotation): + tracking_results = recognize(camera.configs, annotation) + add_recognized_objs(self.con, tracking_results, self.start_time, camera.id) - def retrieve_bbox(self, query: Query = None, world_id: str = ""): + def retrieve_bbox(self, query: Query = None, camera_id: str = ""): bbox = Table(BBOX_TABLE) - q = SnowflakeQuery.from_(bbox).select("*").where(bbox.worldId == world_id) + q = SnowflakeQuery.from_(bbox).select("*").where(bbox.cameraId == camera_id) return query + q if query else q # UNION - def retrieve_traj(self, query: Query = None, world_id: str = ""): + def retrieve_traj(self, query: Query = None, camera_id: str = ""): traj = Table(TRAJ_TABLE) - q = SnowflakeQuery.from_(traj).select("*").where(traj.worldId == world_id) + q = SnowflakeQuery.from_(traj).select("*").where(traj.cameraId == camera_id) return query + q if query else q # UNION def get_bbox(self, query: Query): @@ -331,51 +379,6 @@ def get_video(self, query, cams, boxed): get_video(fetched_meta, cams, self.start_time, boxed) -if __name__ == "__main__": - # Ingest the camera to the world - c1 = create_camera("cam1", 120) - c2 = create_camera("cam2", 150) - - db = Database() - db.insert_bbox_traj( - world_id="myworld", camera_node=c1, recognition_area=BoundingBox(0, 50, 50, 100) - ) - db.insert_bbox_traj( - world_id="myworld2", camera_node=c2, recognition_area=BoundingBox(0, 50, 50, 100) - ) - - q = db.retrieve_traj(world_id="myworld2") - db.cur.execute(q.get_sql()) - res = db.cur.fetchall() - print(res) - - # db.insert_cam("1", 5, "1") - # db.insert_cam("2", 3, "2") - # db.insert_cam("wid-1", c1) - # - # q = db.retrieve_cam(world_id="wid-1") - # res = db.execute_get_query(q) - # print(res) - # # w1 = world() - # q = "" - # camera_node = Camera(cam_id, point, ratio, video_file, metadata_id, lens) - # # w2 = w1.add_camera({cam_id: "1", size: 5, world_id: "1"}) - # q = db.concat_with(query = q, world_id = "1") - # print(q) - - # # w3 = w2.predicate(cam.size < 4) - # q = db.nest_from(query = q, condition = "query.size < 4") - # print(q) - - # # w4 = w3.add_camera({cam_id: "2", size: 3, world_id: "2"}) - # q = db.concat_with(query = q, world_id = "2") - # print(q) - - # # w5 = w4.predicate(cam.size < 6) - # q = db.nest_from(query = q, condition = "query.size < 6") - # print(q) - - # # w6 = w5.get_camera() - # res = db.execute_get_query(query = q) - - # print(list(res)) +Database.insert_bbox_traj.comparators = { + 'annotation': lambda df: df[0].equals(df[1]) +} \ No newline at end of file diff --git a/apperception/new_util.py b/apperception/new_util.py index f40c5ccd..6c9f3023 100644 --- a/apperception/new_util.py +++ b/apperception/new_util.py @@ -1,6 +1,14 @@ import ast +import datetime import os +import time +from typing import Dict, List, Tuple +from tracked_object import TrackedObject +from box import Box +from camera_config import CameraConfig +from pyquaternion import Quaternion +import numpy as np import lens import point import uncompyle6 @@ -9,6 +17,7 @@ get_video_roi) from world_executor import (create_transform_matrix, reformat_fetched_world_coords, world_to_pixel) +from scenic_util import bbox_to_data3d, convert_timestamps, join def create_camera(cam_id, fov): @@ -192,3 +201,211 @@ def compile_lambda(pred): assert isinstance(right.right, ast.Num) y_range.append(right.right.n) return x_range, y_range + + +def recognize(camera_configs: List[CameraConfig], annotation): + annotation = annotation.head(500) + annotations: Dict[str, TrackedObject] = {} + # sample_token_to_frame_num: Dict[str, str] = {} + # for config in camera_configs: + # if config.frame_id not in sample_token_to_frame_num: + # sample_token_to_frame_num[config.frame_id] = [] + # sample_token_to_frame_num[config.frame_id].append(config.frame_num) + + # for a in annotation.itertuples(index=False): + # sample_token = a.sample_token + # if sample_token not in sample_token_to_frame_num: + # continue + # frame_nums = sample_token_to_frame_num[sample_token] + # item_id = a.instance_token + # if item_id not in annotations: + # annotations[item_id] = TrackedObject(a.category, [], []) + + # box = Box(a.translation, a.size, Quaternion(a.rotation)) + + # corners = box.corners() + + # bbox = np.transpose(corners[:, [3, 7]]) + # # print(sample_token, item_id) + # # print(set(frame_nums)) + # for frame_num in set(frame_nums): + # # TODO: fix this: why are there duplicates + # annotations[item_id].bboxes.append(bbox) + # annotations[item_id].frame_num.append(int(frame_num)) + # break + + # for item_id in annotations: + # frame_num = np.array(annotations[item_id].frame_num) + # bboxes = np.array(annotations[item_id].bboxes) + + # index = frame_num.argsort() + + # annotations[item_id].frame_num = frame_num[index].tolist() + # annotations[item_id].bboxes = bboxes[index, :, :] + + # print(item_id, len(annotations[item_id].frame_num) == len(set(annotations[item_id].frame_num))) + for img_file in camera_configs: + # get bboxes and categories of all the objects appeared in the image file + sample_token = img_file.frame_id + frame_num = img_file.frame_num + all_annotations = annotation[annotation["sample_token"] == sample_token] + # camera_info = {} + # camera_info['cameraTranslation'] = img_file['camera_translation'] + # camera_info['cameraRotation'] = img_file['camera_rotation'] + # camera_info['cameraIntrinsic'] = np.array(img_file['camera_intrinsic']) + # camera_info['egoRotation'] = img_file['ego_rotation'] + # camera_info['egoTranslation'] = img_file['ego_translation'] + + for _, ann in all_annotations.iterrows(): + item_id = ann["instance_token"] + if item_id not in annotations: + # annotations[item_id] = {"bboxes": [], "frame_num": []} + # annotations[item_id]["object_type"] = ann["category"] + annotations[item_id] = TrackedObject(ann["category"], [], []) + + box = Box(ann["translation"], ann["size"], Quaternion(ann["rotation"])) + + corners = box.corners() + + # if item_id == '6dd2cbf4c24b4caeb625035869bca7b5': + # # print("corners", corners) + # # transform_box(box, camera_info) + # # print("transformed box: ", box.corners()) + # # corners_2d = box.map_2d(np.array(camera_info['cameraIntrinsic'])) + # corners_2d = transformation(box.center, camera_info) + # print("2d_corner: ", corners_2d) + # overlay_bbox("v1.0-mini/samples/CAM_FRONT/n015-2018-07-24-11-22-45+0800__CAM_FRONT__1532402927612460.jpg", corners_2d) + + bbox = [corners[:, 1], corners[:, 7]] + annotations[item_id].bboxes.append(bbox) + annotations[item_id].frame_num.append(int(frame_num)) + + print("Recognization done, saving to database......") + return annotations + + +def add_recognized_objs( + conn, + formatted_result: Dict[str, TrackedObject], + start_time: datetime.datetime, + camera_id: str +): + for item_id in formatted_result: + object_type = formatted_result[item_id].object_type + recognized_bboxes = np.array(formatted_result[item_id].bboxes) + tracked_cnt = formatted_result[item_id].frame_num + + top_left = recognized_bboxes[:, 0, :] + bottom_right = recognized_bboxes[:, 1, :] + + obj_traj = [] + for i in range(len(top_left)): + current_tl = top_left[i] + current_br = bottom_right[i] + obj_traj.append([current_tl.tolist(), current_br.tolist()]) + + bboxes_to_postgres( + conn, + item_id, + object_type, + "default_color", + start_time, + tracked_cnt, + obj_traj, + camera_id, + ) + + +def bboxes_to_postgres( + conn, + item_id: str, + object_type: str, + color: str, + start_time: datetime.datetime, + timestamps: List[int], + bboxes: List[List[List[float]]], + camera_id: str, +): + converted_bboxes = [bbox_to_data3d(bbox) for bbox in bboxes] + pairs = [] + deltas = [] + for meta_box in converted_bboxes: + pairs.append(meta_box[0]) + deltas.append(meta_box[1:]) + postgres_timestamps = convert_timestamps(start_time, timestamps) + insert_general_trajectory( + conn, item_id, object_type, color, postgres_timestamps, bboxes, pairs, camera_id + ) + # print(f"{item_id} saved successfully") + + +# Insert general trajectory +def insert_general_trajectory( + conn, + item_id: str, + object_type: str, + color: str, + postgres_timestamps: List[str], + bboxes: List[ + List[List[float]] + ], # TODO: should be ((float, float, float), (float, float, float))[] + pairs: List[Tuple[float, float, float]], + camera_id: str, +): + # Creating a cursor object using the cursor() method + cursor = conn.cursor() + + # Inserting bboxes into Bbox table + insert_bbox_trajectories_builder = [] + min_tl = np.full(3, np.inf) + max_br = np.full(3, np.NINF) + + traj_centroids = [] + + prevTimestamp = None + for timestamp, (tl, br), current_point in zip(postgres_timestamps, bboxes, pairs): + if prevTimestamp == timestamp: + continue + prevTimestamp = timestamp + min_tl = np.minimum(tl, min_tl) + max_br = np.maximum(br, max_br) + + # Insert bbox + insert_bbox_trajectories_builder.append( + f""" + INSERT INTO General_Bbox (itemId, cameraId, trajBbox) + VALUES ( + '{item_id}', + '{camera_id}', + STBOX 'STBOX ZT( + ({join([*tl, timestamp])}), + ({join([*br, timestamp])}) + )' + ); + """ + ) + + # Construct trajectory + traj_centroids.append(f"POINT Z ({join(current_point, ' ')})@{timestamp}") + + # Insert the item_trajectory separately + insert_trajectory = f""" + INSERT INTO Item_General_Trajectory (itemId, cameraId, objectType, color, trajCentroids, largestBbox) + VALUES ( + '{item_id}', + '{camera_id}', + '{object_type}', + '{color}', + '{{{', '.join(traj_centroids)}}}', + STBOX 'STBOX Z( + ({join(min_tl)}), + ({join(max_br)}) + )' + ); + """ + + cursor.execute(insert_trajectory) + cursor.execute("".join(insert_bbox_trajectories_builder)) + + # Commit your changes in the database + conn.commit() diff --git a/apperception/new_world.py b/apperception/new_world.py index 770d2847..706c0e05 100644 --- a/apperception/new_world.py +++ b/apperception/new_world.py @@ -11,18 +11,16 @@ from typing import Any, Dict, List, Optional, Set, Tuple import cv2 +from camera_config import CameraConfig import dill as pickle import matplotlib import matplotlib.pyplot as plt import numpy as np import yaml -from bounding_box import WHOLE_FRAME, BoundingBox -from lens import Lens from new_db import Database -from point import Point -from video_context import Camera +from camera import Camera -from apperception.new_util import compile_lambda +from new_util import compile_lambda matplotlib.use("Qt5Agg") print("get backend", matplotlib.get_backend()) @@ -42,7 +40,7 @@ class Type(IntEnum): class World: # all worlds share a db instance - db = Database(reset=True) + db = Database() camera_nodes: Dict[str, Camera] = {} _parent: Optional[World] @@ -135,13 +133,10 @@ def select_by_range(self, cam_id, x_range: Tuple[float, float], z_range: Tuple[f x1=x_min, y1=float("-inf"), z1=z_min, x2=x_max, y2=float("inf"), z2=z_max ) - def recognize(self, cam_id: str, recognition_area: BoundingBox = WHOLE_FRAME): - assert cam_id in World.camera_nodes - - camera_node = World.camera_nodes[cam_id] - node1 = self._insert_bbox_traj(camera_node=camera_node, recognition_area=recognition_area) - node2 = node1._retrieve_bbox(world_id=node1._world_id) - node3 = node2._retrieve_traj(world_id=node1._world_id) + def recognize(self, camera: Camera, annotation): + node1 = self._insert_bbox_traj(camera=camera, annotation=annotation) + node2 = node1._retrieve_bbox(camera_id=camera.id) + node3 = node2._retrieve_traj(camera_id=camera.id) return node3 def get_video(self, cam_ids: List[str] = [], boxed: bool = False): @@ -259,26 +254,14 @@ def filter_pred_relative_to_type(self, pred: Function): type="camera", ) - def add_camera( - self, - cam_id: str, - location: Point, - ratio: float, - video_file: str, - metadata_identifier: str, - lens: Lens, - ): + def add_camera(self, camera: Camera): """ 1. For update method, we create two nodes: the first node will write to the db, and the second node will retrieve from the db 2. For the write node, never double write. (so we use done flag) ... -> [write] -> [retrive] -> ... """ - # TODO: should let the user add a Camera instead of its fields - camera_node = Camera(cam_id, location, ratio, video_file, metadata_identifier, lens) - World.camera_nodes[cam_id] = camera_node - - node1 = self._insert_camera(camera_node=camera_node) - node2 = node1._retrieve_camera(world_id=node1._world_id) + node1 = self._insert_camera(camera=camera) + node2 = node1._retrieve_camera(camera_id=camera.id) return node2 def interval(self, start, end): @@ -330,38 +313,36 @@ def get_time(self): self.db.get_time, )._execute_from_root(Type.BBOX) - def _insert_camera(self, camera_node: Camera): + def _insert_camera(self, camera: Camera): return derive_world( self, {Type.CAM}, self.db.insert_cam, - # does not pass in world_id because we want to use the world_id of the deriving world - camera_node=camera_node, + camera=camera, ) - def _retrieve_camera(self, world_id: str): + def _retrieve_camera(self, camera_id: str): return derive_world( self, {Type.CAM}, self.db.retrieve_cam, - world_id=world_id, + camera_id=camera_id, ) - def _insert_bbox_traj(self, camera_node: Camera, recognition_area: BoundingBox): + def _insert_bbox_traj(self, camera: Camera, annotation): return derive_world( self, {Type.TRAJ, Type.BBOX}, self.db.insert_bbox_traj, - # does not pass in world_id because we want to use the world_id of the deriving world - camera_node=camera_node, - recognition_area=recognition_area, + camera=camera, + annotation=annotation, ) - def _retrieve_bbox(self, world_id: str): - return derive_world(self, {Type.BBOX}, self.db.retrieve_bbox, world_id=world_id) + def _retrieve_bbox(self, camera_id: str): + return derive_world(self, {Type.BBOX}, self.db.retrieve_bbox, camera_id=camera_id) - def _retrieve_traj(self, world_id: str): - return derive_world(self, {Type.TRAJ}, self.db.retrieve_traj, world_id=world_id) + def _retrieve_traj(self, camera_id: str): + return derive_world(self, {Type.TRAJ}, self.db.retrieve_traj, camera_id=camera_id) def _execute_from_root(self, type: Type): nodes: list[World] = [] @@ -384,13 +365,16 @@ def _execute_from_root(self, type: Type): continue # treat update method differently elif node.fn == self.db.insert_cam or node.fn == self.db.insert_bbox_traj: + print('execute:', node.fn.__name__) if not node.done: node._execute() node._done = True node._update_log_file() else: + print('execute:', node.fn.__name__) # print(query) query = node._execute(query=query) + print('done execute node') res = query return res @@ -469,6 +453,7 @@ def _update_log_file(self): } ) ) + pass def empty_world(name: str) -> World: @@ -498,9 +483,9 @@ def _empty_world(name: str) -> World: def derive_world(parent: World, types: set[Type], fn: Any, **kwargs) -> World: - world = _derive_world_from_file(parent, types, fn, **kwargs) - if world is not None: - return world + # world = _derive_world_from_file(parent, types, fn, **kwargs) + # if world is not None: + # return world return _derive_world(parent, types, fn, **kwargs) @@ -582,8 +567,8 @@ def split_filename(filename: str) -> Tuple[str, datetime.datetime, str]: DUMPED_EMPTY_DICT = pickle.dumps({}) -def double_equal(a, b): - return a == b +def double_equal(a: Tuple[Any, Any]): + return a[0] == a[1] def op_matched( @@ -605,10 +590,11 @@ def op_matched( return False cmps = fn.comparators if hasattr(fn, "comparators") else {} - return all( - key in f_kwargs and cmps.get(key, double_equal)(f_kwargs[key], kwargs[key]) + a = all( + key in f_kwargs and cmps.get(key, double_equal)((f_kwargs[key], kwargs[key])) for key in kwargs ) + return a def format_content(content: dict[str, Any]) -> dict[str, Any]: diff --git a/apperception/object_tracker_yolov4_deepsort.py b/apperception/object_tracker_yolov4_deepsort.py index b13fd0c2..000f5b3b 100644 --- a/apperception/object_tracker_yolov4_deepsort.py +++ b/apperception/object_tracker_yolov4_deepsort.py @@ -216,7 +216,7 @@ def yolov4_deepsort_video_track(video_file: str, recognition_area: BoundingBox = formatted_result[item_id].bboxes.append( BoundingBox(int(bbox[0]), int(bbox[1]), int(bbox[2]), int(bbox[3])) ) - formatted_result[item_id].tracked_cnt.append(frame_num) + formatted_result[item_id].frame_num.append(frame_num) else: break diff --git a/apperception/object_tracker_yolov5_deepsort.py b/apperception/object_tracker_yolov5_deepsort.py index b2763830..3ad5ae0e 100644 --- a/apperception/object_tracker_yolov5_deepsort.py +++ b/apperception/object_tracker_yolov5_deepsort.py @@ -179,7 +179,7 @@ def detect(opt: YoloV5Opt): formatted_result[item_id] = TrackedObject(object_type=names[c]) formatted_result[item_id].bboxes.append(bboxes) - formatted_result[item_id].tracked_cnt.append(frame_idx) + formatted_result[item_id].frame_num.append(frame_idx) return formatted_result diff --git a/apperception/scenic_util.py b/apperception/scenic_util.py index e92b51eb..4136627e 100644 --- a/apperception/scenic_util.py +++ b/apperception/scenic_util.py @@ -108,7 +108,7 @@ def create_or_insert_camera_table(conn, world_name, camera): insert_camera( conn, world_name, - fetch_camera_config(camera.scenic_scene_name, camera.object_recognition.sample_data), + fetch_camera_config(camera.id, camera.object_recognition.sample_data), ) return CREATE_CAMERA_SQL @@ -520,6 +520,7 @@ def fetch_camera(conn, scene_name, frame_num): cameraId = '{scene_name}' AND frameNum IN ({",".join(map(str, frame_num))}); """ + print(query) cursor.execute(query) return cursor.fetchall() @@ -583,14 +584,14 @@ def import_tables(conn): # Helper function to convert the timestam to the timestamp formula pg-trajectory uses -def convert_timestamps(start_time, timestamps): +def convert_timestamps(start_time: datetime.datetime, timestamps: Iterable[int]): return [str(start_time + datetime.timedelta(seconds=t)) for t in timestamps] # Helper function to convert trajectory to centroids -def bbox_to_data3d(bbox): +def bbox_to_data3d(bbox: List[List[float]]): """ Compute the center, x, y, z delta of the bbox """ diff --git a/apperception/tracked_object.py b/apperception/tracked_object.py index ce891f19..2bda8293 100644 --- a/apperception/tracked_object.py +++ b/apperception/tracked_object.py @@ -1,11 +1,32 @@ +import numpy as np from dataclasses import dataclass, field -from typing import List - -from bounding_box import BoundingBox - +from typing import List, Union @dataclass class TrackedObject: object_type: str - bboxes: List[BoundingBox] = field(default_factory=list) - tracked_cnt: List[int] = field(default_factory=list) + bboxes: Union[List[np.ndarray], np.ndarray] = field(default_factory=list) + frame_num: List[int] = field(default_factory=list) + + def __eq__(self, other) -> bool: + return ( + isinstance(other, TrackedObject) and + self.object_type == other.object_type and + self.frame_num == other.frame_num and + np.array_equal(np.array(self.bboxes), np.array(other.bboxes)) + ) + + def equal(self, other) -> bool: + if not isinstance(other, TrackedObject) or self.object_type != other.object_type: + return False + + s_frame_num = np.array(self.frame_num) + o_frame_num = np.array(other.frame_num) + + s_indices = s_frame_num.argsort() + o_indices = o_frame_num.argsort() + + return ( + np.array_equal(s_frame_num[s_indices], o_frame_num[o_indices]) and + np.array_equal(np.array(self.bboxes)[s_indices], np.array(other.bboxes)[o_indices]) + ) diff --git a/apperception/video_context.py b/apperception/video_context.py index e17c8bc6..7a1f6918 100644 --- a/apperception/video_context.py +++ b/apperception/video_context.py @@ -9,8 +9,8 @@ @dataclass class Camera: - def __init__(self, scenic_scene_name): - self.scenic_scene_name = scenic_scene_name + def __init__(self, id: str): + self.id = id # Contain objects that still have yet to be added to the backend # If user calls recognize, those items will have already been diff --git a/apperception/video_context_executor.py b/apperception/video_context_executor.py index 7d178969..10feaa33 100644 --- a/apperception/video_context_executor.py +++ b/apperception/video_context_executor.py @@ -48,7 +48,7 @@ def visit_camera(self, camera_node: Camera): return camera_sql def visit_obj_rec(self, camera_node, object_rec_node): - cam_id = camera_node.scenic_scene_name + cam_id = camera_node.id start_time = self.current_context.start_time diff --git a/apperception/video_util.py b/apperception/video_util.py index 067dae4b..72c48165 100644 --- a/apperception/video_util.py +++ b/apperception/video_util.py @@ -310,7 +310,7 @@ def add_recognized_objs( recognized_bboxes = np.array( [bbox.to_tuples() for bbox in formatted_result[item_id].bboxes] ) - tracked_cnt = formatted_result[item_id].tracked_cnt + tracked_cnt = formatted_result[item_id].frame_num top_left = np.vstack((recognized_bboxes[:, 0, 0], recognized_bboxes[:, 0, 1])) if default_depth: top_left_depths = np.ones(len(recognized_bboxes)) diff --git a/benchmarks/new-api-scenic.ipynb b/benchmarks/new-api-scenic.ipynb new file mode 100644 index 00000000..e5bc3d91 --- /dev/null +++ b/benchmarks/new-api-scenic.ipynb @@ -0,0 +1,164 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": null, + "id": "408d6be8", + "metadata": {}, + "outputs": [], + "source": [ + "%cd .." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e582655f", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "import sys\n", + "import json\n", + "import pandas as pd\n", + "import cv2\n", + "import psycopg2\n", + "import pickle\n", + "sys.path.append(os.path.join(os.getcwd(),\"apperception\"))\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "51e7b5ce", + "metadata": {}, + "outputs": [], + "source": [ + "from new_world import empty_world, World\n", + "from camera import Camera" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f6156ab5", + "metadata": {}, + "outputs": [], + "source": [ + "World.db.reset()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0915d633", + "metadata": {}, + "outputs": [], + "source": [ + "# conn = psycopg2.connect(database=\"mobilitydb\", user=\"docker\", password=\"docker\", host=\"localhost\", port=25432)\n", + "# import_tables(conn)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "4641c93d", + "metadata": {}, + "outputs": [], + "source": [ + "name = 'ScenicWorld' # world name\n", + "units = 'metrics' # world units\n", + "user_data_dir = os.path.join(\"v1.0-mini\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b5458d32", + "metadata": {}, + "outputs": [], + "source": [ + "with open('df_sample_data.pickle', \"rb\") as f:\n", + " df_sample_data = pickle.loads(f.read())\n", + "with open('df_annotation.pickle', \"rb\") as f:\n", + " df_annotation = pickle.loads(f.read())" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "43aaae6f", + "metadata": {}, + "outputs": [], + "source": [ + "world = empty_world(name=name)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "46d2d6f7", + "metadata": {}, + "outputs": [], + "source": [ + "from camera_config import fetch_camera_config\n", + "# scenes = [\"scene-0061\", \"scene-0103\",\"scene-0553\", \"scene-0655\", \"scene-0757\", \"scene-0796\", \"scene-0916\", \"scene-1077\", \"scene-1094\", \"scene-1100\"]\n", + "scenes = [\"scene-0061\"]\n", + "for scene in scenes:\n", + " config = fetch_camera_config(scene, df_sample_data)\n", + " camera = Camera(config=config, id=scene)\n", + " world = world.add_camera(camera)\n", + " df_config = df_sample_data[df_sample_data['scene_name'] == scene][['sample_token']]\n", + " df_ann = df_annotation.join(df_config.set_index('sample_token'), on='sample_token', how='inner')\n", + " world = world.recognize(camera, df_ann)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c11fd0c6", + "metadata": {}, + "outputs": [], + "source": [ + "car_trajectories = world.filter_traj_type(object_type='vehicle.car').get_traj()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9a372800", + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a1d4ec08", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.7.12" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} From 5817c7c059ad329cf1af97c284d1e9caacce3cf9 Mon Sep 17 00:00:00 2001 From: Github Actions Bot Date: Sun, 13 Mar 2022 09:40:12 +0000 Subject: [PATCH 4/4] Automated lint and format fixes --- apperception/camera.py | 3 ++- apperception/camera_config.py | 6 +++--- apperception/new_db.py | 39 +++++++++++++++++----------------- apperception/new_util.py | 22 ++++++++----------- apperception/new_world.py | 11 ++++------ apperception/tracked_object.py | 19 +++++++++-------- 6 files changed, 48 insertions(+), 52 deletions(-) diff --git a/apperception/camera.py b/apperception/camera.py index 2acdef73..f6670a5e 100644 --- a/apperception/camera.py +++ b/apperception/camera.py @@ -1,9 +1,10 @@ +import uuid from dataclasses import dataclass from typing import List, Optional -import uuid from camera_config import CameraConfig + @dataclass class Camera: id: str diff --git a/apperception/camera_config.py b/apperception/camera_config.py index e6bed4d6..c3b50d05 100644 --- a/apperception/camera_config.py +++ b/apperception/camera_config.py @@ -4,6 +4,7 @@ Float3 = Tuple[float, float, float] Float4 = Tuple[float, float, float, float] + @dataclass(frozen=True) class CameraConfig: frame_id: str @@ -35,6 +36,5 @@ def fetch_camera_config(scene_name: str, sample_data): ego_rotation=frame.ego_rotation, timestamp=frame.timestamp, ) - for frame - in all_frames.itertuples(index=False) - ] \ No newline at end of file + for frame in all_frames.itertuples(index=False) + ] diff --git a/apperception/new_db.py b/apperception/new_db.py index e3e87623..34cfa8c3 100644 --- a/apperception/new_db.py +++ b/apperception/new_db.py @@ -2,14 +2,13 @@ from typing import Tuple import psycopg2 -from bounding_box import BoundingBox -from lens import PinholeLens -from new_util import create_camera, get_video, video_fetch_reformat, add_recognized_objs, recognize +from camera import Camera +from new_util import (add_recognized_objs, get_video, recognize, + video_fetch_reformat) from pypika import Column, CustomFunction, Table # https://github.com/kayak/pypika/issues/553 # workaround. because the normal Query will fail due to mobility db from pypika.dialects import Query, SnowflakeQuery -from camera import Camera CAMERA_TABLE = "Cameras" TRAJ_TABLE = "Item_General_Trajectory" @@ -26,7 +25,7 @@ def __init__(self): # The start time of the database access object self.start_time = datetime.datetime(2021, 6, 8, 7, 10, 28) - + def reset(self): self._create_camera_table() self._create_item_general_trajectory_table() @@ -76,7 +75,7 @@ def _create_general_bbox_table(self): def _create_item_general_trajectory_table(self): # drop old - q1 = 'DROP TABLE IF EXISTS Item_General_Trajectory CASCADE;' + q1 = "DROP TABLE IF EXISTS Item_General_Trajectory CASCADE;" # create new q2 = """ @@ -95,25 +94,30 @@ def _create_item_general_trajectory_table(self): self.cur.execute(q1) self.cur.execute(q2) self.con.commit() - + def _create_index(self): - self.cur.execute(""" + self.cur.execute( + """ CREATE INDEX IF NOT EXISTS traj_idx ON Item_General_Trajectory USING GiST(trajCentroids); - """) - self.cur.execute(""" + """ + ) + self.cur.execute( + """ CREATE INDEX IF NOT EXISTS item_idx ON General_Bbox(itemId); - """) - self.cur.execute(""" + """ + ) + self.cur.execute( + """ CREATE INDEX IF NOT EXISTS traj_bbox_idx ON General_Bbox USING GiST(trajBbox); - """) + """ + ) self.con.commit() - def insert_cam(self, camera: Camera): values = [ f"""( @@ -128,8 +132,7 @@ def insert_cam(self, camera: Camera): ARRAY{config.ego_rotation}, '{config.timestamp}' )""" - for config - in camera.configs + for config in camera.configs ] self.cur.execute( @@ -379,6 +382,4 @@ def get_video(self, query, cams, boxed): get_video(fetched_meta, cams, self.start_time, boxed) -Database.insert_bbox_traj.comparators = { - 'annotation': lambda df: df[0].equals(df[1]) -} \ No newline at end of file +Database.insert_bbox_traj.comparators = {"annotation": lambda df: df[0].equals(df[1])} diff --git a/apperception/new_util.py b/apperception/new_util.py index 6c9f3023..346ca991 100644 --- a/apperception/new_util.py +++ b/apperception/new_util.py @@ -1,23 +1,22 @@ import ast import datetime import os -import time from typing import Dict, List, Tuple -from tracked_object import TrackedObject -from box import Box -from camera_config import CameraConfig -from pyquaternion import Quaternion -import numpy as np import lens +import numpy as np import point import uncompyle6 +from box import Box +from camera_config import CameraConfig +from pyquaternion import Quaternion +from scenic_util import bbox_to_data3d, convert_timestamps, join +from tracked_object import TrackedObject from video_context import Camera from video_util import (convert_datetime_to_frame_num, get_video_box, get_video_roi) from world_executor import (create_transform_matrix, reformat_fetched_world_coords, world_to_pixel) -from scenic_util import bbox_to_data3d, convert_timestamps, join def create_camera(cam_id, fov): @@ -211,7 +210,7 @@ def recognize(camera_configs: List[CameraConfig], annotation): # if config.frame_id not in sample_token_to_frame_num: # sample_token_to_frame_num[config.frame_id] = [] # sample_token_to_frame_num[config.frame_id].append(config.frame_num) - + # for a in annotation.itertuples(index=False): # sample_token = a.sample_token # if sample_token not in sample_token_to_frame_num: @@ -233,7 +232,7 @@ def recognize(camera_configs: List[CameraConfig], annotation): # annotations[item_id].bboxes.append(bbox) # annotations[item_id].frame_num.append(int(frame_num)) # break - + # for item_id in annotations: # frame_num = np.array(annotations[item_id].frame_num) # bboxes = np.array(annotations[item_id].bboxes) @@ -285,10 +284,7 @@ def recognize(camera_configs: List[CameraConfig], annotation): def add_recognized_objs( - conn, - formatted_result: Dict[str, TrackedObject], - start_time: datetime.datetime, - camera_id: str + conn, formatted_result: Dict[str, TrackedObject], start_time: datetime.datetime, camera_id: str ): for item_id in formatted_result: object_type = formatted_result[item_id].object_type diff --git a/apperception/new_world.py b/apperception/new_world.py index 706c0e05..4c2f1e61 100644 --- a/apperception/new_world.py +++ b/apperception/new_world.py @@ -11,15 +11,13 @@ from typing import Any, Dict, List, Optional, Set, Tuple import cv2 -from camera_config import CameraConfig import dill as pickle import matplotlib import matplotlib.pyplot as plt import numpy as np import yaml -from new_db import Database from camera import Camera - +from new_db import Database from new_util import compile_lambda matplotlib.use("Qt5Agg") @@ -365,16 +363,16 @@ def _execute_from_root(self, type: Type): continue # treat update method differently elif node.fn == self.db.insert_cam or node.fn == self.db.insert_bbox_traj: - print('execute:', node.fn.__name__) + print("execute:", node.fn.__name__) if not node.done: node._execute() node._done = True node._update_log_file() else: - print('execute:', node.fn.__name__) + print("execute:", node.fn.__name__) # print(query) query = node._execute(query=query) - print('done execute node') + print("done execute node") res = query return res @@ -453,7 +451,6 @@ def _update_log_file(self): } ) ) - pass def empty_world(name: str) -> World: diff --git a/apperception/tracked_object.py b/apperception/tracked_object.py index 2bda8293..276769e5 100644 --- a/apperception/tracked_object.py +++ b/apperception/tracked_object.py @@ -1,7 +1,9 @@ -import numpy as np from dataclasses import dataclass, field from typing import List, Union +import numpy as np + + @dataclass class TrackedObject: object_type: str @@ -10,12 +12,12 @@ class TrackedObject: def __eq__(self, other) -> bool: return ( - isinstance(other, TrackedObject) and - self.object_type == other.object_type and - self.frame_num == other.frame_num and - np.array_equal(np.array(self.bboxes), np.array(other.bboxes)) + isinstance(other, TrackedObject) + and self.object_type == other.object_type + and self.frame_num == other.frame_num + and np.array_equal(np.array(self.bboxes), np.array(other.bboxes)) ) - + def equal(self, other) -> bool: if not isinstance(other, TrackedObject) or self.object_type != other.object_type: return False @@ -26,7 +28,6 @@ def equal(self, other) -> bool: s_indices = s_frame_num.argsort() o_indices = o_frame_num.argsort() - return ( - np.array_equal(s_frame_num[s_indices], o_frame_num[o_indices]) and - np.array_equal(np.array(self.bboxes)[s_indices], np.array(other.bboxes)[o_indices]) + return np.array_equal(s_frame_num[s_indices], o_frame_num[o_indices]) and np.array_equal( + np.array(self.bboxes)[s_indices], np.array(other.bboxes)[o_indices] )