Separate the metrics variables from non-trainable variables. #910
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
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## main #910 +/- ##
==========================================
- Coverage 76.82% 73.25% -3.58%
==========================================
Files 329 329
Lines 31427 31434 +7
Branches 6112 6114 +2
==========================================
- Hits 24144 23027 -1117
- Misses 5719 6893 +1174
+ Partials 1564 1514 -50
Flags with carried forward coverage won't be shown. Click here to find out more.
☔ View full report in Codecov by Sentry. |
fchollet
reviewed
Sep 18, 2023
| @@ -602,6 +603,17 @@ def non_trainable_weights(self): | |||
| return self.weights | |||
| return [v for v in self.weights if not v.trainable] | |||
|
|
|||
| @property | |||
| def metrics_variables(self): | |||
There was a problem hiding this comment.
Does this work with compiled metrics? We should add support for that in the Trainer.
There was a problem hiding this comment.
The trainer override this method in
keras-core/keras_core/trainers/trainer.py
Line 233 in a465816
There was a problem hiding this comment.
Yes, the existing test in trainer does cover that.
There was a problem hiding this comment.
Eg in
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Outdated suggestions cannot be applied.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.
As discussed in #897, we separate the metrics related variables from non-trainable variables, so that we can properly leverage the jax memory donation.
This will also allow us to skip the saving for metrics variables during checkpoint/savemodel