Skip to content

Pull requests: keras-team/keras

New pull request New
77 Open 8,845 Closed
77 Open 8,845 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

feat: Add Grain dataset support for keras.layers.TextVectorization.adapt() size:M
#22402 opened Mar 12, 2026 by dataCenter430 Loading…
1
@gbaned
2
Add optimized GRU for JAX backend size:M
#22401 opened Mar 11, 2026 by MarcosAsh Loading…
@gbaned
6
Replace Python asserts with self.assertXXX in unit tests. size:L
#22400 opened Mar 11, 2026 by hertschuh Loading…
@gbaned
5
Add cuDNN LSTM for JAX backend size:M
#22399 opened Mar 11, 2026 by MarcosAsh Loading…
@gbaned
8
Adding torch support for Data Parallel size:L
#22397 opened Mar 11, 2026 by buildwithsuhana • Draft
@gbaned
6
Adding Torch support for Model Parallel
#22394 opened Mar 10, 2026 by buildwithsuhana • Draft
6
Fix LayerNormalization rms_scaling to use mean of squares instead of variance size:S
#22393 opened Mar 10, 2026 by rstar327 Loading…
1
@gbaned
4
Replace cast+multiply masking with ops.where to reduce memory size:M
#22392 opened Mar 10, 2026 by rstar327 Loading…
1
@gbaned
11
[OpenVINO backend] Support for nanarmax and nanar min size:M
#22383 opened Mar 8, 2026 by andersendsa Loading…
1
@gbaned
8
Add optional Gated Attention size:L stat:awaiting response from contributor
#22372 opened Mar 6, 2026 by pass-lin Loading…
@gbaned
27
Add fmod to keras.ops.numpy size:M stat:awaiting response from contributor
#22369 opened Mar 6, 2026 by rstar327 Loading…
@gbaned
7
Don't skip attention tests on Torch anymore. size:M
#22367 opened Mar 6, 2026 by hertschuh Loading…
1
@gbaned
2
[OpenVINO backend] Support for numpy.eigh size:S stat:awaiting response from contributor
#22363 opened Mar 5, 2026 by andersendsa Loading…
@gbaned
10
Fix #20598 — Custom models/layers with sublayers in nested lists fail to save/load weights size:M
#22362 opened Mar 5, 2026 by pctablet505 • Draft
@gbaned
5
Fix #20429 — Attention layer hardcodes 3-D shape assumptions, breaking N-D inputs size:S
#22361 opened Mar 5, 2026 by pctablet505 • Draft
@gbaned
3
Fix #20322 — Weight save/load fails when model uses a custom subclass of a built-in layer (e.g. LSTM) size:M
#22360 opened Mar 5, 2026 by pctablet505 • Draft
@gbaned
5
Fix/20260 layers property setter awaiting review size:M
#22359 opened Mar 5, 2026 by pctablet505 Loading…
1
@gbaned
2
Fix/20256 early stopping reset size:S
#22358 opened Mar 5, 2026 by pctablet505 • Draft
@gbaned
5
Fix TensorBoard callback step counter never updating size:S
#22357 opened Mar 5, 2026 by pctablet505 • Draft
@gbaned
5
Fix #19127 — Improve error message for named inputs mismatch in Functional model size:S
#22356 opened Mar 5, 2026 by pctablet505 • Draft
@gbaned
4
[OpenVINO backend] Support for numpy.det using Faddeev-LeVerrier algorithm size:M
#22355 opened Mar 5, 2026 by andersendsa Loading…
@gbaned
5
Add multi-backend benchmark comparison support size:L
#22346 opened Mar 4, 2026 by MarcosAsh Loading…
@gbaned
7
Fix #22162: Add input validation for PReLU and LeakyReLU layers size:S
#22344 opened Mar 4, 2026 by thakoreh Loading…
1
@gbaned
5
[OpenVINO] Fix array(), maximum(), minimum(), reshape() and split() ops in the numpy suite size:M stat:awaiting response from contributor
#22343 opened Mar 4, 2026 by goyaladitya05 • Draft
1
@gbaned
14
Add sobel_edges to keras.ops.image size:L
#22334 opened Mar 3, 2026 by rstar327 Loading…
1
@gbaned
7
ProTip! Updated in the last three days: updated:>2026-03-08.