search

thomasmesnard / DeepMind-Teaching-Machines-to-Read-and-Comprehend

Implementation of "Teaching Machines to Read and Comprehend" proposed by Google DeepMind
MIT License
408 stars 104 forks source link

A bug in the model? #5

Open violet-zct opened 8 years ago

violet-zct commented 8 years ago

I found that when representing the context word (concatenating the forward and backward rnn), the backward rnn is not reversed to match the sequence order of the forward rnn.

rabberbock commented 8 years ago

@violet-zct Were you able to use the model to actually make predictions?

violet-zct commented 8 years ago

Yes, I have replied to one of the posts concerning this issue. Create your test data stream and add another DataStreamMonitoring to the extensions in train.py.

2016-05-24 3:03 GMT+08:00 rabberbock notifications@github.com:

@violet-zct https://github.com/violet-zct Were you able to use the model to actually make predictions?

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/thomasmesnard/DeepMind-Teaching-Machines-to-Read-and-Comprehend/issues/5#issuecomment-221059766