TensorBoard/Wandb/optuna/raytune integration improvements. #7935
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
…tune support, with minor modifications to trainer core code.
madlag
force-pushed
the
optuna_integration_improve
branch
from
3a7a0ee
to
5191210
Compare
sgugger
reviewed
Oct 20, 2020
There was a problem hiding this comment.
I don't think changing the metrics names in the logs is a good idea as there are other components relying on them. Renaming the keys inside the TensorBoardCallback to make tensorboard happier seems fine but outside of it, I would prefer to avoid any breaking change.
madlag
force-pushed
the
optuna_integration_improve
branch
from
e339f01
to
55fa620
Compare
madlag
force-pushed
the
optuna_integration_improve
branch
from
55fa620
to
d57d45f
Compare
sgugger
approved these changes
Oct 21, 2020
fabiocapsouza
pushed a commit
to fabiocapsouza/transformers
that referenced
this pull request
Nov 15, 2020
…ce#7935) Improved TensorBoard and Wandb integration, as well as optuna and ray/tune support, with minor modifications to trainer core code.
fabiocapsouza
added a commit
to fabiocapsouza/transformers
that referenced
this pull request
Nov 15, 2020
…uggingface#7935)" This reverts commit e07eb0d.
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Outdated suggestions cannot be applied.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.
Improves TensorBoard logging by grouping train / eval metrics as it is usually done in TensorBoard.
Improves TensorBoard/optuna model hyper-parameters logging.
Improves optuna and Ray/tune integration, and provides model hyper-parameter naming.
Test (and sample code) is provided in test_trainer.TrainerHyperParameterIntegrationTest .
Some more work may be need to harmonize metrics naming for eval / train, as the "eval_" prefix used is not very convenient, using a "eval/" prefix would be more foolproof, and consistent with TensorBoard usage, but it would break quite some code, and so may be done in a separate PR.