Add backwards-compatibility logic for model progress tracker #2468
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
justinxzhao
reviewed
Sep 9, 2022
Comment on lines
528
to
555
| for stat in ("improvement", "increase_batch_size", "learning_rate_reduction"): | ||
| assert f"last_{stat}_epoch" not in new_model_progress | ||
| assert f"last_{stat}_steps" in new_model_progress | ||
| assert ( | ||
| new_model_progress[f"last_{stat}_steps"] | ||
| == old_model_progress[f"last_{stat}_epoch"] * old_model_progress["batch_size"] | ||
| ) | ||
|
|
||
| assert "tune_checkpoint_num" in new_model_progress | ||
|
|
||
| assert "vali_metrics" not in new_model_progress | ||
| assert "validation_metrics" in new_model_progress | ||
|
|
||
| metric = new_model_progress["validation_metrics"]["combined"]["loss"][0] | ||
| assert len(metric) == 3 | ||
| assert metric[-1] == 0.59 | ||
|
|
||
| # Verify that we don't make changes to already-valid model progress dicts. | ||
| # To do so, we modify the batch size value and re-run the upgrade on the otherwise-valid `new_model_progress` dict. | ||
| new_model_progress["batch_size"] = 1 | ||
| unchanged_model_progress = upgrade_model_progress(new_model_progress) | ||
|
|
||
| for stat in ("improvement", "increase_batch_size", "learning_rate_reduction"): | ||
| assert unchanged_model_progress[f"last_{stat}_steps"] == new_model_progress[f"last_{stat}_steps"] | ||
|
|
||
| unchanged_metric = unchanged_model_progress["validation_metrics"]["combined"]["loss"][0] | ||
| new_metric = new_model_progress["validation_metrics"]["combined"]["loss"][0] | ||
| assert unchanged_metric == new_metric |
There was a problem hiding this comment.
I find that it's easier to read a single check for equality on the entire dictionary, wholesale. What do you think?
There was a problem hiding this comment.
Agreed, just need to specify an expected config dict.
tgaddair
approved these changes
Sep 10, 2022
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.
TrainerMetrictuples, also with estimated epoch and step values.