Pull requests: keras-team/keras
Pull requests list
Allow per-variable optimizer, add DispatchOptimizer.
size:L
#21196
opened Apr 21, 2025 by
cantonios
Loading…
Adding functional CompositeLayer
awaiting review
size:XL
#21099
opened Mar 27, 2025 by
martin-gorner
Loading…
Enable ROCm devices to correctly use keras.layers.CuDNNGRU and keras.layers.CuDNNLSTM
awaiting review
size:M
#21443
opened Jul 1, 2025 by
JoeNan1
Loading…
Add TF support for mixed optional inputs (None/tensor) from generator in model.fit/evaluate/predict
size:M
#21609
opened Aug 22, 2025 by
neo-alex
Loading…
Fix Discretization layer graph mode bug
size:S
#21514
opened Jul 25, 2025 by
divyashreepathihalli
Loading…
[keras/src/print.py] Implement backend-specialised
print function
size:S
#21344
opened Jun 2, 2025 by
SamuelMarks
Loading…
Non cumulative model.predict
awaiting review
size:M
#21673
opened Sep 17, 2025 by
pctablet505
Loading…
Introducing Parameter Sharding and Torch backend for Tensor Parallelism
size:XL
#21724
opened Oct 7, 2025 by
buildwithsuhana
Loading…
avoid failure when docstrings have been stripped (python -OO)
size:M
#21477
opened Jul 14, 2025 by
JohannesBuchner
Loading…
Adding Tensor_layout for Tensor parallelism for Autosharding
size:L
#21792
opened Oct 28, 2025 by
buildwithsuhana
Loading…
Fix misleading documentation for smart_resize backend_module parameter
size:XS
stat:awaiting response from contributor
#21722
opened Oct 5, 2025 by
Flakes342
Loading…
Fix: handle 3D input for channels_first in preprocess_input() #21749
size:M
stat:awaiting response from contributor
#21754
opened Oct 17, 2025 by
utsab345
Loading…
Fix Nested Metrics Handling in CompileMetrics
size:L
stat:awaiting response from contributor
#21761
opened Oct 20, 2025 by
Dimios45
Loading…
5 tasks done
Fix #21105: JAX Backend - Custom Gradient Variable Capture Issue in keras.ops.custom_gradient
size:L
stat:awaiting response from contributor
#21783
opened Oct 26, 2025 by
Mayankvlog
Loading…
Fix: convert deque to list before tf.transpose in keras.ops.quantile
size:S
stat:awaiting response from contributor
#21784
opened Oct 26, 2025 by
Junead04
Loading…
Fix Muon optimizer to use variable.name instead of deprecated variable.path
size:M
stat:awaiting response from contributor
#21854
opened Nov 16, 2025 by
MalyalaKarthik66
Loading…
Fix: Remove deprecated .path access in Muon optimizer for TF 2.16+ compatibility
size:M
stat:awaiting response from contributor
#21797
opened Oct 29, 2025 by
utsab345
Loading…
📝docs: clarify shuffle behavior and example in PyDataset
size:S
stat:awaiting response from contributor
#21847
opened Nov 13, 2025 by
maitry63
Loading…
feat: Add a predict_proba method on SKLearnClassifier
awaiting review
size:XS
#21556
opened Aug 7, 2025 by
divakaivan
Loading…
Fix: correct effective crop_to_aspect_ratio logic and lint issues
size:M
stat:awaiting response from contributor
#21779
opened Oct 25, 2025 by
utsab345
Loading…
Fix ops.tile shape inference issue on TensorFlow backend
size:M
#21860
opened Nov 19, 2025 by
ssam18
Loading…
Fix custom layer parameters ending in _mask (#21154)
size:L
stat:awaiting response from contributor
#21811
opened Nov 1, 2025 by
Mayankvlog
Loading…
ProTip!
Add no:assignee to see everything that’s not assigned.