François Chollet comments

Results 342 comments of


                                            François Chollet

Selecting Metal (MPS) as the GPU in MacOS (torch backend)

With this approach, you should be able to write a custom training loop, right? Does that work?

Selecting Metal (MPS) as the GPU in MacOS (torch backend)

One thing we could do is introduce a torch data iterator for `fit()` that moves the data to the default torch device.

Add AdamWScheduleFree

Thanks for the suggestion. It looks very doable. We'd have to wait to see it gain traction before adding it to the core API though. Are you willing to work...

Feature request: fold/unfold ops

Thanks for the suggestions. Do these ops have any equivalent in JAX or in NumPy?

Layer "dense" expects 1 input(s), but it received 2 input tensors error while loading a keras model

Try with `keras-nightly`, this error should be fixed already.

RNN not compatible with XLA (TF backend)

We should disable jit_compile in auto mode if backend is TF and there's a LSTM or GRU layer and cuDNN is usable for them

as_list() is not defined on an unknown TensorShape.

It looks like your dataset is yielding tensors that have no shape (i.e. not even a rank), and that isn't supported by augmentation layers. Fix: make sure your dataset yields...

scikit-learn wrappers in keras

Thanks for the suggestion! I think these wrappers do provide value to the community. > I'd be happy to help (as a scikit-learn maintainer) if y'all decide to add them...

scikit-learn wrappers in keras

Ok, that's a good argument. If you'd like to open a PR, we'll review it.

Convert the handwriting recognition example to Keras 3.

Did you take a look at the data on local disk and checked everything looked good?