Update saving_lib.py to properly convert from h5py to numpy array. #21371
Conversation
This file contains hidden or 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 ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #21371 +/- ##
==========================================
+ Coverage 82.70% 82.72% +0.02%
==========================================
Files 565 565
Lines 54894 54906 +12
Branches 8517 8521 +4
==========================================
+ Hits 45398 45419 +21
+ Misses 7415 7399 -16
- Partials 2081 2088 +7
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
Hey @pctablet505 Previously, I simply added an else block for this but I have updated it to your solution. |
Thanks for including it. |
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Outdated suggestions cannot be applied.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.
What does this PR do?
Fixes an incompatibility when loading model weights in environments where TensorFlow's NumPy behavior is enabled (
tf.experimental.numpy.experimental_enable_numpy_behavior). Previously, some variables loaded as h5py datasets were not converted to numpy arrays, causing downstream errors during model construction and weight assignment, especially for layers likeReversibleEmbeddingandEinsumDense.This update ensures that all h5py datasets are converted to numpy arrays before further processing in
saving_lib.py.Why is this needed?
When using TensorFlow's NumPy behavior, type promotion and array handling can differ, leading to incompatibilities if weights remain as h5py objects. This change guarantees consistent array types regardless of backend or NumPy mode.
Related Issues/PRs