Decode mode produces same generated summary for all articles

[theHimanshuShekhar]

If running the model in decode mode using the flag it shows the same generated summary for all the input articles. Any help would be amazing.

[DengYangyong]

I meet the sample problem that all the decoded results of different examples are the same no matter I set rl_training=Flase or True, and I have trained the model for about 300000 steps and the loss has stopped failling.

Have you solved the issue?

[theHimanshuShekhar]

i went ahead with the pytorch implementation and abandoned trying to fix this. I will leave this issue up in @amrzaki ever looks at this. It might help him fix the issue