diff --git a/your-code/lab_imbalance.ipynb b/your-code/lab_imbalance.ipynb index a3a5359..4ef0464 100644 --- a/your-code/lab_imbalance.ipynb +++ b/your-code/lab_imbalance.ipynb @@ -28,11 +28,350 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": 35, "metadata": {}, "outputs": [], "source": [ - "# Your code here" + "import pandas as pd\n", + "import numpy as np" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "metadata": {}, + "outputs": [], + "source": [ + "# Your response here\n", + "data = pd.read_csv(\"data.csv\")" + ] + }, + { + "cell_type": "code", + "execution_count": 37, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
steptypeamountnameOrigoldbalanceOrgnewbalanceOrignameDestoldbalanceDestnewbalanceDestisFraudisFlaggedFraud
01PAYMENT9839.64C1231006815170136.0160296.36M19797871550.00.000
11PAYMENT1864.28C166654429521249.019384.72M20442822250.00.000
21TRANSFER181.00C1305486145181.00.00C5532640650.00.010
31CASH_OUT181.00C840083671181.00.00C3899701021182.00.010
41PAYMENT11668.14C204853772041554.029885.86M12307017030.00.000
\n", + "
" + ], + "text/plain": [ + " step type amount nameOrig oldbalanceOrg newbalanceOrig \\\n", + "0 1 PAYMENT 9839.64 C1231006815 170136.0 160296.36 \n", + "1 1 PAYMENT 1864.28 C1666544295 21249.0 19384.72 \n", + "2 1 TRANSFER 181.00 C1305486145 181.0 0.00 \n", + "3 1 CASH_OUT 181.00 C840083671 181.0 0.00 \n", + "4 1 PAYMENT 11668.14 C2048537720 41554.0 29885.86 \n", + "\n", + " nameDest oldbalanceDest newbalanceDest isFraud isFlaggedFraud \n", + "0 M1979787155 0.0 0.0 0 0 \n", + "1 M2044282225 0.0 0.0 0 0 \n", + "2 C553264065 0.0 0.0 1 0 \n", + "3 C38997010 21182.0 0.0 1 0 \n", + "4 M1230701703 0.0 0.0 0 0 " + ] + }, + "execution_count": 37, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 38, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(6362620, 11)" + ] + }, + "execution_count": 38, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data.shape" + ] + }, + { + "cell_type": "code", + "execution_count": 39, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
steptypeamountnameOrigoldbalanceOrgnewbalanceOrignameDestoldbalanceDestnewbalanceDestisFraudisFlaggedFraud
5870199403PAYMENT17509.95C482191795303892.08286382.13M3446485300.000.0000
4193018305CASH_OUT164344.44C174247915048329.600.00C1268033679204610.35368954.7900
31419316CASH_IN182877.50C16256408973824909.554007787.05C21417765863283310.232965018.0900
5790096401TRANSFER136487.99C1030520262052.760.00C9378850862215463.062351951.0500
3911997284CASH_IN217344.37C4955728894917976.085135320.46C148782608211181782.3910964438.0100
\n", + "
" + ], + "text/plain": [ + " step type amount nameOrig oldbalanceOrg \\\n", + "5870199 403 PAYMENT 17509.95 C482191795 303892.08 \n", + "4193018 305 CASH_OUT 164344.44 C1742479150 48329.60 \n", + "314193 16 CASH_IN 182877.50 C1625640897 3824909.55 \n", + "5790096 401 TRANSFER 136487.99 C103052026 2052.76 \n", + "3911997 284 CASH_IN 217344.37 C495572889 4917976.08 \n", + "\n", + " newbalanceOrig nameDest oldbalanceDest newbalanceDest isFraud \\\n", + "5870199 286382.13 M344648530 0.00 0.00 0 \n", + "4193018 0.00 C1268033679 204610.35 368954.79 0 \n", + "314193 4007787.05 C2141776586 3283310.23 2965018.09 0 \n", + "5790096 0.00 C937885086 2215463.06 2351951.05 0 \n", + "3911997 5135320.46 C1487826082 11181782.39 10964438.01 0 \n", + "\n", + " isFlaggedFraud \n", + "5870199 0 \n", + "4193018 0 \n", + "314193 0 \n", + "5790096 0 \n", + "3911997 0 " + ] + }, + "execution_count": 39, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "n = 100000\n", + "data_sample = data.sample(n=n, axis=0)\n", + "\n", + "data_sample.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(100000, 11)" + ] + }, + "execution_count": 40, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data_sample.shape" ] }, { @@ -44,7 +383,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 41, "metadata": {}, "outputs": [], "source": [ @@ -60,11 +399,599 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": 42, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "step int64\n", + "type object\n", + "amount float64\n", + "nameOrig object\n", + "oldbalanceOrg float64\n", + "newbalanceOrig float64\n", + "nameDest object\n", + "oldbalanceDest float64\n", + "newbalanceDest float64\n", + "isFraud int64\n", + "isFlaggedFraud int64\n", + "dtype: object" + ] + }, + "execution_count": 42, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Your code here\n", + "data_sample.dtypes" + ] + }, + { + "cell_type": "code", + "execution_count": 43, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "step 0\n", + "type 0\n", + "amount 0\n", + "nameOrig 0\n", + "oldbalanceOrg 0\n", + "newbalanceOrig 0\n", + "nameDest 0\n", + "oldbalanceDest 0\n", + "newbalanceDest 0\n", + "isFraud 0\n", + "isFlaggedFraud 0\n", + "dtype: int64" + ] + }, + "execution_count": 43, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data_sample.isnull().sum()" + ] + }, + { + "cell_type": "code", + "execution_count": 44, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "CASH_OUT 35351\n", + "PAYMENT 33775\n", + "CASH_IN 21904\n", + "TRANSFER 8311\n", + "DEBIT 659\n", + "Name: type, dtype: int64" + ] + }, + "execution_count": 44, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data_sample[\"type\"].value_counts()" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "metadata": {}, + "outputs": [], + "source": [ + "dummies = pd.get_dummies(data_sample[\"type\"])" + ] + }, + { + "cell_type": "code", + "execution_count": 46, "metadata": {}, "outputs": [], "source": [ - "# Your code here\n" + "new_data = pd.concat([data_sample, dummies], axis=1)" + ] + }, + { + "cell_type": "code", + "execution_count": 47, + "metadata": {}, + "outputs": [], + "source": [ + "new_data.drop(\"type\", inplace = True, axis = 1)" + ] + }, + { + "cell_type": "code", + "execution_count": 48, + "metadata": {}, + "outputs": [], + "source": [ + "new_data.drop(\"nameOrig\", inplace = True, axis = 1) #deleted because is a string" + ] + }, + { + "cell_type": "code", + "execution_count": 49, + "metadata": {}, + "outputs": [], + "source": [ + "new_data.drop(\"nameDest\", inplace = True, axis = 1) #deleted because is a string" + ] + }, + { + "cell_type": "code", + "execution_count": 50, + "metadata": {}, + "outputs": [], + "source": [ + "new_data.drop(\"isFlaggedFraud\", inplace = True, axis = 1) #deleted because only has values with 0" + ] + }, + { + "cell_type": "code", + "execution_count": 51, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
stepamountoldbalanceOrgnewbalanceOrigoldbalanceDestnewbalanceDestisFraudCASH_INCASH_OUTDEBITPAYMENTTRANSFER
587019940317509.95303892.08286382.130.000.00000010
4193018305164344.4448329.600.00204610.35368954.79001000
31419316182877.503824909.554007787.053283310.232965018.09010000
5790096401136487.992052.760.002215463.062351951.05000001
3911997284217344.374917976.085135320.4611181782.3910964438.01010000
\n", + "
" + ], + "text/plain": [ + " step amount oldbalanceOrg newbalanceOrig oldbalanceDest \\\n", + "5870199 403 17509.95 303892.08 286382.13 0.00 \n", + "4193018 305 164344.44 48329.60 0.00 204610.35 \n", + "314193 16 182877.50 3824909.55 4007787.05 3283310.23 \n", + "5790096 401 136487.99 2052.76 0.00 2215463.06 \n", + "3911997 284 217344.37 4917976.08 5135320.46 11181782.39 \n", + "\n", + " newbalanceDest isFraud CASH_IN CASH_OUT DEBIT PAYMENT TRANSFER \n", + "5870199 0.00 0 0 0 0 1 0 \n", + "4193018 368954.79 0 0 1 0 0 0 \n", + "314193 2965018.09 0 1 0 0 0 0 \n", + "5790096 2351951.05 0 0 0 0 0 1 \n", + "3911997 10964438.01 0 1 0 0 0 0 " + ] + }, + "execution_count": 51, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "new_data.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 52, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "step int64\n", + "amount float64\n", + "oldbalanceOrg float64\n", + "newbalanceOrig float64\n", + "oldbalanceDest float64\n", + "newbalanceDest float64\n", + "isFraud int64\n", + "CASH_IN uint8\n", + "CASH_OUT uint8\n", + "DEBIT uint8\n", + "PAYMENT uint8\n", + "TRANSFER uint8\n", + "dtype: object" + ] + }, + "execution_count": 52, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "new_data.dtypes" + ] + }, + { + "cell_type": "code", + "execution_count": 53, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
stepamountoldbalanceOrgnewbalanceOrigoldbalanceDestnewbalanceDestisFraudCASH_INCASH_OUTDEBITPAYMENTTRANSFER
step1.0000000.022357-0.013393-0.0133250.0262080.0243800.0355860.007531-0.0183340.0012950.0087590.005080
amount0.0223571.000000-0.004457-0.0090610.2714240.4334450.073067-0.010418-0.003892-0.022992-0.1927820.359373
oldbalanceOrg-0.013393-0.0044571.0000000.9988340.0649780.0419420.0077650.508080-0.201436-0.021589-0.189429-0.081529
newbalanceOrig-0.013325-0.0090610.9988341.0000000.0662940.041518-0.0098590.528501-0.211654-0.021998-0.193733-0.086935
oldbalanceDest0.0262080.2714240.0649780.0662941.0000000.978410-0.0051000.0713190.0854540.011058-0.2242430.126101
newbalanceDest0.0243800.4334450.0419420.0415180.9784101.0000000.0020680.0322360.0935390.008347-0.2327540.186032
isFraud0.0355860.0730670.007765-0.009859-0.0051000.0020681.000000-0.0187360.017057-0.002881-0.0252650.042662
CASH_IN0.007531-0.0104180.5080800.5285010.0713190.032236-0.0187361.000000-0.391622-0.043135-0.378211-0.159447
CASH_OUT-0.018334-0.003892-0.201436-0.2116540.0854540.0935390.017057-0.3916221.000000-0.060228-0.528088-0.222632
DEBIT0.001295-0.022992-0.021589-0.0219980.0110580.008347-0.002881-0.043135-0.0602281.000000-0.058165-0.024521
PAYMENT0.008759-0.192782-0.189429-0.193733-0.224243-0.232754-0.025265-0.378211-0.528088-0.0581651.000000-0.215008
TRANSFER0.0050800.359373-0.081529-0.0869350.1261010.1860320.042662-0.159447-0.222632-0.024521-0.2150081.000000
\n", + "
" + ], + "text/plain": [ + " step amount oldbalanceOrg newbalanceOrig \\\n", + "step 1.000000 0.022357 -0.013393 -0.013325 \n", + "amount 0.022357 1.000000 -0.004457 -0.009061 \n", + "oldbalanceOrg -0.013393 -0.004457 1.000000 0.998834 \n", + "newbalanceOrig -0.013325 -0.009061 0.998834 1.000000 \n", + "oldbalanceDest 0.026208 0.271424 0.064978 0.066294 \n", + "newbalanceDest 0.024380 0.433445 0.041942 0.041518 \n", + "isFraud 0.035586 0.073067 0.007765 -0.009859 \n", + "CASH_IN 0.007531 -0.010418 0.508080 0.528501 \n", + "CASH_OUT -0.018334 -0.003892 -0.201436 -0.211654 \n", + "DEBIT 0.001295 -0.022992 -0.021589 -0.021998 \n", + "PAYMENT 0.008759 -0.192782 -0.189429 -0.193733 \n", + "TRANSFER 0.005080 0.359373 -0.081529 -0.086935 \n", + "\n", + " oldbalanceDest newbalanceDest isFraud CASH_IN CASH_OUT \\\n", + "step 0.026208 0.024380 0.035586 0.007531 -0.018334 \n", + "amount 0.271424 0.433445 0.073067 -0.010418 -0.003892 \n", + "oldbalanceOrg 0.064978 0.041942 0.007765 0.508080 -0.201436 \n", + "newbalanceOrig 0.066294 0.041518 -0.009859 0.528501 -0.211654 \n", + "oldbalanceDest 1.000000 0.978410 -0.005100 0.071319 0.085454 \n", + "newbalanceDest 0.978410 1.000000 0.002068 0.032236 0.093539 \n", + "isFraud -0.005100 0.002068 1.000000 -0.018736 0.017057 \n", + "CASH_IN 0.071319 0.032236 -0.018736 1.000000 -0.391622 \n", + "CASH_OUT 0.085454 0.093539 0.017057 -0.391622 1.000000 \n", + "DEBIT 0.011058 0.008347 -0.002881 -0.043135 -0.060228 \n", + "PAYMENT -0.224243 -0.232754 -0.025265 -0.378211 -0.528088 \n", + "TRANSFER 0.126101 0.186032 0.042662 -0.159447 -0.222632 \n", + "\n", + " DEBIT PAYMENT TRANSFER \n", + "step 0.001295 0.008759 0.005080 \n", + "amount -0.022992 -0.192782 0.359373 \n", + "oldbalanceOrg -0.021589 -0.189429 -0.081529 \n", + "newbalanceOrig -0.021998 -0.193733 -0.086935 \n", + "oldbalanceDest 0.011058 -0.224243 0.126101 \n", + "newbalanceDest 0.008347 -0.232754 0.186032 \n", + "isFraud -0.002881 -0.025265 0.042662 \n", + "CASH_IN -0.043135 -0.378211 -0.159447 \n", + "CASH_OUT -0.060228 -0.528088 -0.222632 \n", + "DEBIT 1.000000 -0.058165 -0.024521 \n", + "PAYMENT -0.058165 1.000000 -0.215008 \n", + "TRANSFER -0.024521 -0.215008 1.000000 " + ] + }, + "execution_count": 53, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "new_data.corr()" ] }, { @@ -76,11 +1003,34 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 57, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "0.9985\n", + "Training data accuracy was 0.9987\n" + ] + } + ], "source": [ - "# Your code here" + "# Your code here\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.linear_model import LogisticRegression\n", + "from sklearn.metrics import accuracy_score\n", + "\n", + "features = new_data.drop(\"isFraud\", axis = 1)\n", + "target = new_data[\"isFraud\"]\n", + "\n", + "X_train, X_test, y_train, y_test = train_test_split(features, target, test_size = 0.20, random_state = 0)\n", + "\n", + "log_reg = LogisticRegression()\n", + "log_reg.fit(X_train, y_train)\n", + "\n", + "print(log_reg.score(X_test, y_test))\n", + "print(f\"Training data accuracy was {log_reg.score(X_train, y_train)}\")" ] }, { @@ -92,11 +1042,28 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 59, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "0.5383219853506562\n", + "Training data accuracy was 0.9989275131167832\n" + ] + } + ], "source": [ - "# Your code here" + "# Your code here\n", + "#Extreme Gradient Boosting\n", + "import xgboost\n", + "\n", + "xgb_reg = xgboost.XGBRegressor(max_depth = 5, n_estimators = 500)\n", + "xgb_reg.fit(X_train, y_train)\n", + "\n", + "print(xgb_reg.score(X_test, y_test))\n", + "print(f\"Training data accuracy was {xgb_reg.score(X_train, y_train)}\")" ] }, { @@ -108,11 +1075,11 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ - "# Your response here" + "# Your response here: Logistic logisitc regression worked better. Extreme Gradient Boosting is overfitting." ] }, { @@ -125,7 +1092,7 @@ ], "metadata": { "kernelspec": { - "display_name": "Python 3", + "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, @@ -139,7 +1106,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.8" + "version": "3.10.9" } }, "nbformat": 4,