Make use of prefilled cache in beam_search when predict_batch_with_aux.prompt_with_targets is set to True
#1388
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
![copybara-service[bot]](https://avatars.githubusercontent.com/in/44061?s=60&v=4){kind=small}
copybara-service
bot
force-pushed
the
test_559256044
branch
from
2ce0ef0 to
aac9b3f
Compare
copybara-service
bot
force-pushed
the
test_559256044
branch
3 times, most recently
from
a710dea to
864e95e
Compare
copybara-service
bot
force-pushed
the
test_559256044
branch
2 times, most recently
from
340b634 to
de7577e
Compare
copybara-service
bot
force-pushed
the
test_559256044
branch
from
de7577e to
34f1e74
Compare
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Outdated suggestions cannot be applied.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.
Make use of prefilled cache in beam_search when
predict_batch_with_aux.prompt_with_targetsis set to TrueWithout this change, the targets that are actually prompts will still be decoded AR style in order to build the cache, the logits are thrown away.
With this change, and setting
EncoderDecoder.predict_batch_with_aux.prompt_with_targettoTruewe instead skip over the given prompts all together, speeding up the decode.