More tests to Trainer #6699
Merged
More tests to Trainer #6699
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
sgugger
commented
Aug 24, 2020
| @@ -77,6 +77,7 @@ jobs: | |||
| - v0.3-torch_and_tf-{{ checksum "setup.py" }} | |||
| - v0.3-{{ checksum "setup.py" }} | |||
| - run: pip install --upgrade pip | |||
| - run: pip install git+https://github.com/huggingface/nlp | |||
There was a problem hiding this comment.
For now, using nlp master branch to test the integration with Trainer (we need some unreleased features). Not sure if we want to keep it this way for the foreseeable future or use the last release later on.
sgugger
commented
Aug 24, 2020
| self.data_collator = data_collator if data_collator is not None else default_data_collator | ||
| self.train_dataset = train_dataset | ||
| self.eval_dataset = eval_dataset | ||
| self.model_init = model_init | ||
| self.compute_metrics = compute_metrics | ||
| self.optimizer, self.lr_scheduler = optimizers | ||
| if model_init is not None and (self.optimizer is not None or self.lr_scheduler is not None): | ||
| raise RuntimeError( | ||
| "Passing a `model_init` is incompatible with providing the `optimizers` argument." |
There was a problem hiding this comment.
Raising an error as a defense mechanism: model_init means each new training starts from scratch, thus needs a clean optimizer/scheduler.
sgugger
commented
Aug 24, 2020
| ) | ||
| self.hp_search_backend = backend | ||
|
|
||
| if self.model_init is None: | ||
| raise RuntimeError( | ||
| "To use hyperparameter search, you need to pass your model through a model_init function." |
There was a problem hiding this comment.
Defense mechanism: you can't use HP search if the model is not reinitialized at each training.
Codecov Report
@@ Coverage Diff @@
## master #6699 +/- ##
==========================================
- Coverage 79.44% 79.01% -0.43%
==========================================
Files 156 156
Lines 28386 28388 +2
==========================================
- Hits 22551 22432 -119
- Misses 5835 5956 +121
Continue to review full report at Codecov.
|
LysandreJik
approved these changes
Aug 25, 2020
Comment on lines
+213
to
+214
| # Seed must be set before instantiating the model when using model | ||
| set_seed(self.args.seed) |
Zigur
pushed a commit
to Zigur/transformers
that referenced
this pull request
Oct 26, 2020
* More tests to Trainer * Add warning in the doc
fabiocapsouza
pushed a commit
to fabiocapsouza/transformers
that referenced
this pull request
Nov 15, 2020
* More tests to Trainer * Add warning in the doc
fabiocapsouza
added a commit
to fabiocapsouza/transformers
that referenced
this pull request
Nov 15, 2020
This reverts commit 28e8214.
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.
While doing see, realized there were some problems with the seed (in particular for HP search) so added a few tests of that too.