Generated Review problem

[akhileshkumargangwar]

I want to ask how data is splitted into postive and negative reviews.When I checked manually many reviews whose score is 1 or 2 has been assigned to positive folder (discriminator_train/positive).Logically it is negative review. After adversial training in this (train_sample_generated/7epoch_step2_temp_positive/000012.txt) file generated review is just copy of original given review. Orginal Input Review-- {"review": "i wasnt thrilled with the taste of the food compared to how pricy it is . i would have enjoyed a juicy steak somewhere else . however i do like the romance of fondue", "score": "3"} Generated Review in file 000012.txt---{"label": "1", "example": "i would have enjoyed a juicy steak somewhere else . however i do like the romance of fondue"}

It has just reduced one sentence. Is it like that.

[jingjingxupku]

"discriminator_train/positive" stores the real data. "discriminator_train/negative" stores the data generated by our code. It is important to note that "positive/negative" means the positive samples and negative samples for discriminator, rather than favourable/disapproving comments.

[akhileshkumargangwar]

Thank You, please give some hint about my second question.

[jingjingxupku]

train_sample_generated/7epoch_step2_temp_positive/000012.txt is not the generated data. It is the copy of original reviews to show what the real-life reviews should be. If you want to see the generated data, please open the folder "train_sample_generated/7epoch_step2_temp_negative/000012.txt".

[akhileshkumargangwar]

Thank You, Your code is very good and has used text data as input. I am confused with positive, negative terms. Thanks to make your code publicly available. May you please give some description in readme about which generated folders contains what.Many folders has been generated and how to find corresponding input to generated output. Your code is good compare to other available code like SeqGAN and others. It will be helpful for novice researchers.

[jingjingxupku]

I will add more folder descriptions in readme.md soon. Thank you for your suggestion!

[akhileshkumargangwar]

Please add folder description

[jingjingxupku]

Sorry for the delay. I will upload the folder description in 24 hours.

[akhileshkumargangwar]

Thanks for updating Readme

[mambalong]

@akhileshkumargangwar How long does it take to run the code? I run the code by running the "run.sh" file. It has been 24 hours now, it is still running. And I am using a GPU.