keras-team keras issues

keras-team / keras

Deep Learning for humans

http://keras.io/

Apache License 2.0

62.12k stars 19.49k forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Problem with GRU simulation

#20535 kokhazade opened 25 minutes ago
0
Fix `BaseOptimizer` to handle a mix of `tf.Variable` and `KerasVariable`

#20534 james77777778 opened 8 hours ago
1
Allow EarlyStopping to be reused between multiple `fit`s.

#20533 hertschuh opened 11 hours ago
1
Add inner op

#20532 IMvision12 opened 13 hours ago
1
AttributeError: module 'keras_nlp' has no attribute 'models'

#20531 iwqculrbud opened 22 hours ago
1
Raise an error when `save_weights` or `load_weights` is called on an unbuilt model

#20530 james77777778 closed 20 hours ago
1
Cannot access accuracy in results with keras 3

#20529 tanwarsh opened 1 day ago
2
Replace `np.iinfo` and `np.finfo` with `ml_dtypes`

#20528 james77777778 closed 1 day ago
1
Update example code in keras_activations_linear file

#20527 LavanyaKV1234 closed 1 day ago
1
implement transform_bounding_boxes for random_zoom

#20526 shashaka opened 1 day ago
2
Fix incorrect argument name and update description in RNN documentation

#20525 rameshdange5191 closed 1 day ago
2
Accuracy is lost after save_weights/load_weights

#20524 Pandaaaa906 opened 2 days ago
3
Fixing example code for CategoricalCrossentropy in losses.py file

#20523 tilakrayal closed 2 days ago
1
Fix and update GQA tests

#20522 james77777778 closed 2 days ago
1
EarlyStopping instance is not reusable

#20521 maciejbiesek opened 2 days ago
0
Code of Conduct for Community Standards Proposal

#20520 joshuajk7 closed 1 day ago
1
Add Geomspace ops

#20519 IMvision12 closed 2 days ago
1
Need to explicitly specify `x` and `y` instead of using a generator when the model has two inputs.

#20518 danica-zhu opened 2 days ago
0
apply_gradients AttributeError: 'ResourceVariable' object has no attribute 'overwrite_with_gradient'

#20517 andrewl36 opened 2 days ago
3
Make test resilient to spurious warnings.

#20516 hertschuh closed 2 days ago
1
Update example code in hard_sigmoid file

#20515 keerthanakadiri closed 1 day ago
1
Introduce flash attention to GQA and move the flash attention configs to `backend.config`

#20514 james77777778 closed 3 days ago
1
Allow `convert_to_tensor` to take a value with the wrong `dtype` on T…

#20513 hertschuh closed 3 days ago
1
Avoid call to deprecated xla_bridge.get_backend()

#20512 jakevdp closed 3 days ago
1
`MultiHeadAttention._compute_attention_mask()` always returns a bool …

#20511 hertschuh closed 3 days ago
1
Fix tensorflow `_dot_product_attention_xla` and update `enable_flash_attention`

#20510 james77777778 closed 4 days ago
1
Add squareplus activation

#20508 shashaka closed 4 days ago
1
Training discrepancy between Keras 2 and 3

#20507 matemijolovic closed 5 days ago
3
Add exp2 op

#20506 IMvision12 closed 5 days ago
1
Add exp2 op

#20505 IMvision12 closed 6 days ago
1
Add exp2 numpy op

#20504 IMvision12 closed 6 days ago
1
More flexible output_shape computation in keras.layers.MultiHeadAttention

#20503 lcs-crr closed 4 days ago
2
Minor changes to asserts in keras.src.layers.attention.MultiHeadAttention

#20502 lcs-crr closed 6 days ago
0
Fix rendering issue

#20501 sachinprasadhs closed 5 days ago
1
Unpin coverage

#20499 james77777778 closed 6 days ago
1
Add exp2 operator to keras ops

#20498 larschristensen closed 4 days ago
2
Will Keras 3 support intel gpu with oneapi

#20497 HaroldYin1024 closed 5 days ago
7
model.fit() doesn't update loss values properly with a custom loss function in Keras 3

#20496 youmemonk opened 1 week ago
1
Enhance the robustness of the flash attention check

#20495 james77777778 closed 1 week ago
3
Add soft_shrink activation

#20494 shashaka closed 1 week ago
1
is QAT still on the roadmap for keras3?

#20493 matpalm closed 6 days ago
1
Fixing example code for BinaryFocalCrossentropy in losses.py file

#20492 tilakrayal closed 1 week ago
1
implement transform_bounding_boxes for center_crop

#20491 shashaka closed 6 days ago
4
ModelCheckpoint loses .h5 save support, breaking retrocompatibility

#20490 TeoCavi opened 1 week ago
2
Refactor `dot_product_attention` to use flash attention when available

#20489 james77777778 closed 1 week ago
2
Fix `attention_mask` computation in `MultiHeadAttention`

#20488 james77777778 closed 1 week ago
2
Add parameter axis to tversky loss

#20487 jakubxy08 opened 1 week ago
1
Support custom cell/RNN layers with extension types

#20485 Johansmm opened 1 week ago
4
Improve the consistency of the names of initializers

#20484 james77777778 closed 1 week ago
1
Issue loading model containing Dense layer with Identity initializer

#20483 kristoferm94 closed 1 week ago
2