This repository was archived by the owner on Jul 7, 2023. It is now read-only.
Bug fixes in inference and data generation; faster token unescaping #162
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
lukaszkaiser
approved these changes
Jul 18, 2017
Contributor
lukaszkaiser
left a comment
lukaszkaiser
left a comment
There was a problem hiding this comment.
Looks good, thanks Villi.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
2 participants
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Outdated suggestions cannot be applied.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.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR contains three fixes/improvements:
In
generator_utils.py, fixes regression wheretokenizer.token_countswas referred to in the latest release of T2T but this variable/attribute does not exist.In
text_encoder.py, uses a regex to unescape tokens, which is much faster than the previous while loop. In the most common case, the regex matches nothing and the token string is passed through unchanged (into the final underscore cut-off).In
trainer_utils.py, when decoding from a dataset,estimator.predict()is (as of the latest release of T2T) called withas_iterable=False. This requires a change to the decoding loop, so that inputs, targets and outputs are correctly iterated.