reproduction of the results of paper

[Aneureka]

Hi, I'm confused to reproduce the results in the paper. The data and results seemed to be different from the paper... can you help me? (my English is poor...

[yangtiangithub]

May be some data are polished when paper writing. Have you complied the code successfully, I can't even complied it, there are two many errors in the code, I tested it in Centos 7. All the steps are okay except for the last make in code/cnn. How did you make it ?

[liberty1997]

@yangtiangithub , I fixed some bugs/errors by adding a header(#include "fmt/printf.h") in following files: code/nn/include/config.h, code/nn/include/inet.h . Because there are some differences between fmt-4.0 and older release in "format.h". But there are still two errors: "/usr/bin/ld: cannot find -lfmt" , "/usr/bin/ld: cannot find -lwht" . I found libfmt but not libwht.

[liberty1997]

Uh, I made it. ( "run test")

[josephenguehard]

@liberty1997 did you get the same results as in the paper? I was able to run the code (so_run.sh) but got an error rate of around 54.5%

[liberty1997]

@liberty1997 did you get the same results as in the paper? I was able to run the code (so_run.sh) but got an error rate of around 54.5%

I'm sorry, I just helped a senior schoolmate to adjust the whole program, and the final operation was done by her. I don't know what the result is.

[josephenguehard]

Ok thanks!