Search results

Usages of Keras Augmentaiton Layers

Status: Open.

#21954 In keras-team/keras;

· innat opened on Dec 25, 2025

Training stateful RNN fails due to symbolic tensor

type:Bug

Status: Open.

#21942 In keras-team/keras;

· fhchl opened on Dec 19, 2025

Upgrade JAX dependency

dependenciesPull requests that update a dependency filePull requests that update a dependency file

Status: Open.

#21926 In keras-team/keras;

· adamjstewart opened on Dec 14, 2025

<code>jax.errors.ConcretizationTypeError</code> in Splash attention during <code>model.generate</code>

stat:awaiting response from contributor

type:Bug

Status: Open.

#21916 In keras-team/keras;

· pctablet505 opened on Dec 10, 2025

Upgrade <code>ai-edge-litert</code> with new TensorFlow release

dependenciesPull requests that update a dependency filePull requests that update a dependency file

Status: Open.

#21914 In keras-team/keras;

· abheesht17 opened on Dec 10, 2025

Add CLAHE

type:featureThe user is asking for a new feature.The user is asking for a new feature.

Status: Open.

#21898 In keras-team/keras;

· innat opened on Dec 5, 2025

Feature Request: convert built models from channels_first to channels_last and vice versa.

type:featureThe user is asking for a new feature.The user is asking for a new feature.

Status: Open.

#21889 In keras-team/keras;

· DiGyt opened on Dec 3, 2025

Q: Torch memory usage vs Tensorflow

stale

stat:awaiting response from contributor

Status: Open.

#21870 In keras-team/keras;

· teaglin opened on Nov 25, 2025

RandomCrop does nothing at validation/inference

keras-team-review-pendingPending review by a Keras team member.Pending review by a Keras team member.

type:Bug

Status: Open.

#21868 In keras-team/keras;

· ma7555 opened on Nov 24, 2025

Q: is it possible (any examples?) to train two different models on two different GPUs, in different threads of the same process?

stale

stat:awaiting response from contributor

Status: Open.

#21862 In keras-team/keras;

· mw66 opened on Nov 20, 2025

Compiled torch model has much worse performance than non-compiled version.

type:supportUser is asking for help / asking an implementation question. Stackoverflow would be better suited.User is asking for help / asking an implementation question. Stackoverflow would be better suited.

Status: Open.

#21851 In keras-team/keras;

· blechn opened on Nov 14, 2025

Add RandomResizedCrop

type:featureThe user is asking for a new feature.The user is asking for a new feature.

Status: Open.

#21822 In keras-team/keras;

· innat opened on Nov 5, 2025