Join GitHub today
GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.
Sign up#174 fixed that unk_replacement and beam_search work together #271
Conversation
googlebot
commented
Jul 6, 2017
|
Thanks for your pull request. It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). Once you've signed, please reply here (e.g.
|
|
I signed it! |
googlebot
commented
Jul 6, 2017
|
CLAs look good, thanks! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
nem6ishi commentedJul 6, 2017
The issue that unk_replacement and beam_search does not work is solved in this pull request.
The problem is that there is no attension_scores but multiple attension_scores in beam_search which is named "beam_search_output.original_outputs.attention_scores".
So I just added the procedures to pick one of them.
Since it seems that the top of the list is always the best among the beams in beam_search, [:,0,:] of it was taken.
This is my first big pull request and the code does not look beautiful, but I will appreciate if I could get kind instructions for this.