Added logsumexp to backend. #6346
Merged
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
fchollet
reviewed
Apr 22, 2017
keras/backend/tensorflow_backend.py
Outdated
|
|
||
|
|
||
| def logsumexp(x, axis=None, keepdims=False): | ||
| '''Returns `log(sum(exp(x), axis=axis, keepdims=keepdims))` with improved numerical stability. |
There was a problem hiding this comment.
The one-line docstring summary should fit within 80 chars.
Both added functions need a compliant docstring (# Arguments, # Returns).
There was a problem hiding this comment.
Thanks. It is fixed now.
|
LGTM |
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Outdated suggestions cannot be applied.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.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds the method logsumexp(x, axis, keepdims) to the backend.
Tensorflow (resp. Theano) offers an optimization for computing
log(sum(exp(x)))asx_max + log(sum(exp(x - x_max)))that improves numerical stability for large values. It is exposed bytf.reduce_logsumexp(resp. internal graph optimization).Having this functionality simplifies building a Conditional Random Field with Keras (see PR #4621).