Fix Pthon code from https://github.com/facebookresearch/end-to-end-negotiator
Budget: $10 – $100 USD
I try to run script in Python from:
https://github.com/facebookresearch/end-to-end-negotiator
I try in VMware (CentOS 7 and Ubuntu 16.04) and again on real PC (CentOS 7 and Ubuntu 16.04).
Script run without any errors and generate files:
selection_model.th and rnn_model.th which is needed for first Selfplay test where two pretrained models negotiate against each another.
But script don't output generated dialogues (I see only some statistics).
I decompress file: selection_model.th (it's ZIP file) and in directory: archive > data > I found that all 49 files have the same content (only 4 digits):
00 00 C0 FF
so file is corrupted and not trained.
File: rnn_model.th looks OK because in directory archive > data > all files are different.
Your job is to find what is wrong and what I need to do to make it work OK.
https://github.com/facebookresearch/end-to-end-negotiator
I try in VMware (CentOS 7 and Ubuntu 16.04) and again on real PC (CentOS 7 and Ubuntu 16.04).
Script run without any errors and generate files:
selection_model.th and rnn_model.th which is needed for first Selfplay test where two pretrained models negotiate against each another.
But script don't output generated dialogues (I see only some statistics).
I decompress file: selection_model.th (it's ZIP file) and in directory: archive > data > I found that all 49 files have the same content (only 4 digits):
00 00 C0 FF
so file is corrupted and not trained.
File: rnn_model.th looks OK because in directory archive > data > all files are different.
Your job is to find what is wrong and what I need to do to make it work OK.