Keras Add keras.jit

It seems that we need a Keras JIT program.Now that torch.compile, tf.function, and jax.jit can all implement JIT functio...

Keras Does Keras have an op similar to jax.jvp?

I’d like to ask whether Keras offers a backend-agnostic equivalent of jax.jvp; if not, I can contribute an implementatio...

Keras 'jit_compile=True' raises InternalTorchDynamoError with EfficientNetV2 models on torch backend.

As per title, compiling a Keras EfficientNetV2 model using torch backend raises an InternalTorchDynamoError, see below.E...

Keras ValueError: Invalid dtype: property object - ops.prod with layers.Dense

The output of ops.prod doesn't work in layer's argument. To make it work, it's required to cast, i.e. int(ops.prod). Now...

Keras Error with keras.applications.EfficientNetB0()

To reproduce run: python3 -c 'import keras; keras.applications.EfficientNetB0()'python3 -c 'import keras;keras.applicati...

Keras Whether 'jpg' is supported or not in func keras.utils.save_img()

Bug IssueI found an issue on keras.utils.save_img() with arg file_format='jpg', the doc of keras.utils.save_img() shows ...

Keras Unexpected behavior with ops.pad (tested with Tensorflow and JAX backend)

I am encountering a bug with ops.pad and the Tensorflow backend (I am using Keras 3.11.1 and Tensorflow 2.19.0., current...

Keras Support multiple validation sets in Model.fit

Moving this feature request from TF repository https://github.com/tensorflow/tensorflow/issues/38803Are you willing to c...

Keras Keras Policy forces 64-bit integer arithmetic results to int32

Describe the bugThe arithmetic operation A+B, where both A and B are int64 Keras Variables (tested on TensorFlow), unexp...

Keras How Do I Track a List of Objects for Export?

I define an EnsembleModel class that is constructed from a list of other Keras models.class EnsembleModel(keras.Model): ...

Keras Compatibility Matrix for GPU based workflow

Last night I was trying to code some examples with jax backend but couldn't get it working because of cuda and cuddnn is...

Keras Differences in Model Weight Behavior Between Keras 2.15 and Keras 3

I attempted to migrate name-based Keras 2.x weights in .h5 format to Keras 3 by loading the .h5 weights into the model a...

Keras RandAugment incorrect behavior in tf graph mode

In tensorflow graph mode, RandAugment will repeatedly use the same augmentations, which were sampled during graph tracin...

Keras TF = 2.18, Keras= 3.10 Unrecognized keyword arguments passed to MultiHeadAttention: {'return_attention_scores': True}

I have the following block, and I am getting the error "Unrecognized keyword arguments passed to MultiHeadAttention: {'...

Keras Integrate Grain into Keras

@james77777778 Comment From: james77777778https://github.com/keras-team/keras/pull/21494Comment From: github-actions[bot...

Keras Check if model is applicable when compile

Bug IssueThe doc of keras.applications.MobileNetV3Small() shows its description as below:https://github.com/keras-team/k...

Keras Incoherent behaviour in Model.compile between loss & metrics for dict output Models

Hey!I believe there is a bug in keras 3 related to metrics in Model.compile. At the very least there is a significant di...

Keras torch backend: keras.utils.split_dataset requires tensorflow

The keras.utils.split_dataset function requires tensorflow to be installed even tho the docstring states that the datase...

Keras ValueError: Received incompatible devices for jitted computation. (demo_jax_distributed.py)

I get the following error when running examples/demo_jax_distributed.py on a Cloud TPU VM (tpuv2):Traceback (most recent...

Keras Simple RNN for system identification

As I have described in the following issue on Stack Exchange, I fail to get good results using SimpleRNN to capture very...

下一页
.