xvector: Objf and Deriv #6
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
added 3 commits
February 14, 2016 17:19
…. Improving unit tests. Adding CPU version of objf calculation
| } else if (i < j) { | ||
| obfj_terms[scores_index] = K * log(1.0 + exp(L)); | ||
| obfj_derivs[scores_index] = -K / (1.0 + exp(-L)); | ||
| } |
Owner
There was a problem hiding this comment.
if I were you I'd do at this point
else {
objf_terms[scores_index] = 0.0;
objf_derivs[scores_index] = 0.0;
}
this will save you from having to initialize those to zero (i.e. you can size them with kUndefined)
Owner
There was a problem hiding this comment.
Also, please don't assume that the stride of objf_deriv is the same as that of objf. It's against the interface.
danpovey
pushed a commit
that referenced
this pull request
Jan 5, 2018
* OCR: Add IAM corpus with unk decoding support (#3) * Add a new English OCR database 'UW3' * Some minor fixes re IAM corpus * Fix an issue in IAM chain recipes + add a new recipe (#6) * Some fixes based on the pull request review * Various fixes + cleaning on IAM * Fix LM estimation and add extended dictionary + other minor fixes * Add README for IAM * Add output filter for scoring * Fix a bug RE switch to pyhton3 * Add updated results + minor fixes * Remove unk decoding -- gives almost no gain * Add UW3 OCR database * Fix cmd.sh in IAM + fix usages of train/decode_cmd in chain recipes * Various minor fixes on UW3 * Rename iam/s5 to iam/v1 * Add README file for UW3 * Various cosmetic fixes on UW3 scripts * Minor fixes in IAM
danpovey
pushed a commit
that referenced
this pull request
Jan 5, 2018
* OCR: Add IAM corpus with unk decoding support (#3) * Add a new English OCR database 'UW3' * Some minor fixes re IAM corpus * Fix an issue in IAM chain recipes + add a new recipe (#6) * Some fixes based on the pull request review * Various fixes + cleaning on IAM * Fix LM estimation and add extended dictionary + other minor fixes * Add README for IAM * Add output filter for scoring * Fix a bug RE switch to pyhton3 * Add updated results + minor fixes * Remove unk decoding -- gives almost no gain * Add UW3 OCR database * Fix cmd.sh in IAM + fix usages of train/decode_cmd in chain recipes * Various minor fixes on UW3 * Rename iam/s5 to iam/v1 * Add README file for UW3 * Various cosmetic fixes on UW3 scripts * Minor fixes in IAM
danpovey
pushed a commit
that referenced
this pull request
Nov 7, 2019
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Finished cuda-enabled version of objf and deriv calculation.
Improving unit tests.
Added CPU version of objf calculation.