François Chollet
François Chollet
That sounds like something we can try. If we do that, we should introduce a new kind of object (not `__bytes__`) since we need to be backwards compatible with already...
Thanks for the PR -- please take a look at the test failures. Also, we've recently merged another change related to loss management in multi-output models, so the PR would...
> If you'd like I can keep only the test changes. Otherwise, feel free to close. Any increase in test coverage is valuable! Please do.
> It seems that the current codebase is missing tf.VariableSynchronization, as implemented in tf-keras: I thought it was only relevant for MultiWorkerMirroredStrategy, which we don't support in Keras 3. But...
The test appears to be merely flaky.
Thanks for the PR! Can you link to a Colab showing the callback in action (maybe with different backends)?
Please make sure to insert line breaks in log messages so that the logs do not interfere too much with the progress bar printouts.
> First time contributing. Need a little clarification please. Are you familiar with PyTorch Datasets and Dataloaders?
I suggest just removing the arg from the documentation (and signature) entirely. It can just be hidden in `kwargs`.
Thanks for the suggestion! It seems this requires an external library to handle rar files -- it cannot be done with the Python standard library. What's the most popular one?...