Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] The Missing Ingredient in Zero-Shot Neural Machine Translation #1714

Open
wants to merge 13 commits into
base: master
Choose a base branch
from
+135 −29
Open

[WIP] The Missing Ingredient in Zero-Shot Neural Machine Translation #1714

Changes from 1 commit
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
f35b34e
add lambda_cosine, move normalization to compute_loss, adapt stats
francoishernandez Jan 30, 2020
845c989
fix some flake
francoishernandez Jan 30, 2020
86f2c52
Merge branch 'master' of https://github.com/OpenNMT/OpenNMT-py into c…
francoishernandez Feb 3, 2020
84e472f
fix forward tests
francoishernandez Feb 3, 2020
1a6e737
disable sharded loss if lambda_cosine
francoishernandez Feb 3, 2020
e99daaf
fix some flake
francoishernandez Feb 3, 2020
1faea3c
move cosine loss compute to function, fix some args
francoishernandez Feb 4, 2020
f8bc7f4
fix flake
francoishernandez Feb 4, 2020
9d26360
add arxiv link
francoishernandez Feb 7, 2020
306b2e5
broadcast instead of explicitly create ones
francoishernandez Feb 7, 2020
8616b99
return encoder representations only if necessary
francoishernandez Feb 17, 2020
75d645a
roll back tests for model forward
francoishernandez Feb 17, 2020
f53ea4d
fix typo
francoishernandez Feb 17, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
fix some flake
@francoishernandez
francoishernandez committed Feb 3, 2020
commit e99daafae3d36e4250dd6370846c92e1950db8b3
4 changes: 2 additions & 2 deletions onmt/utils/parse.py
View file
Original file line number Diff line number Diff line change
@@ -121,8 +121,8 @@ def validate_train_opts(cls, opt):
"Number of attention_dropout values must match accum_steps values"

assert not(opt.max_generator_batches > 0 and opt.lambda_cosine != 0), \
"-lambda_cosine loss is not implemented for max_generator_batches > 0."

"-lambda_cosine loss is not implemented " \
"for max_generator_batches > 0."

@classmethod
def validate_translate_opts(cls, opt):