run keras compatibility test for sequence loss with v2 behavior. #1374
+79
−53
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Outdated suggestions cannot be applied.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.
I runs in full eager mode + with the tf.function enabled by default for keras layers.
@guillaumekln that might be useful for #1371
get_test_data
is here to replace thesetup
method. I'll convert to pytest the rest of the tests in this file in another pull request.np.testing.assert_allclose(calculated_loss, expected_loss, rtol=1e-6, atol=1e-6)
np.testing.assert_allclose
is more strict than the tf counterpartself.assertAllClose
so I took the rtol and atol from theself.assertAllClose
to have the same behavior. Otherwise the tests were not passing.