Source/Target length distinction #1
Merged
Conversation
This file contains 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
Removed parameter -seq_length New parameters -src_seq_length and -tgt_seq_length
As in the LUA OpenNMT
pltrdy
changed the title
bmccann
added a commit
that referenced
this pull request
Apr 19, 2017
srush
pushed a commit
that referenced
this pull request
Aug 22, 2017
srush
pushed a commit
that referenced
this pull request
Nov 5, 2017
Fix unintended float type for num_block in SRU.py
pltrdy
referenced
this pull request
in pltrdy/OpenNMT-py
May 14, 2018
Fix 'whitespace-only translation handling'
Gldkslfmsd
pushed a commit
to Gldkslfmsd/OpenNMT-py
that referenced
this pull request
Sep 6, 2018
prototyping multi-encoder
saikat107
pushed a commit
to saikat107/OpenNMT-py
that referenced
this pull request
Mar 12, 2019
Fix 'whitespace-only translation handling'
saikat107
pushed a commit
to saikat107/OpenNMT-py
that referenced
this pull request
Mar 12, 2019
small misspelling bug
bhacquin
pushed a commit
to bhacquin/OpenNMT-py
that referenced
this pull request
Jun 11, 2019
bhacquin
pushed a commit
to bhacquin/OpenNMT-py
that referenced
this pull request
Jun 14, 2019
ldeseynes
pushed a commit
to ldeseynes/OpenNMT-py
that referenced
this pull request
Jun 4, 2020
minor changes to make code simpler/more explicit
Closed
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Outdated suggestions cannot be applied.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.
Preprocess parameters
Removed parameter
-seq_lengthNew parameters
-src_seq_lengthand-tgt_seq_lengthTraining speed token/s
In both LUA/PyTorch OpenNMT, the training process prints a speed, in token/sec, but:
This can lead to important differences, especially when src/tgt sequence length are different (e.g. summarization), and therefore lead to false conclusion about performances.
See also: pytoch/example/issue#75