-
Notifications
You must be signed in to change notification settings - Fork 348
Pull requests: tensorflow/model-optimization
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
This change updates the random number generation utility to use tf.uint64 for bitwise operations and constants. The seed generation function (next_seed_fn) has been rewritten to implement a proper 7-bit LFSR (PRBS7) using bitwise operations, replacing the previous power-based approximation.
#1288
opened May 27, 2026 by
copybara-service
Bot
Loading…
The
fully_commutes_with_sum property was incorrectly using sum() on the flattened commuting structure, which does not yield a boolean. Changed to use all() to ensure the property returns True only if all sub-encoders fully commute with sum. Added type assertions in tests to verify the return type is boolean.
#1286
opened May 25, 2026 by
copybara-service
Bot
Loading…
Fix #1272: tensorflow-model-optimization fails to import in official ...
#1276
opened Apr 18, 2026 by
JiwaniZakir
Loading…
docs: clarify quantized Add + tanh fusion limitation in TFLite
#1256
opened Jan 28, 2026 by
Atchayasree03
Loading…
Internal change.
technique:pruning
Regarding tfmot.sparsity.keras APIs and docs
#1169
opened Jan 13, 2025 by
copybara-service
Bot
Loading…
Fix model loading from .keras files in Tensorflow 2.15
technique:qat
Regarding tfmot.quantization.keras (for quantization-aware training) APIs and docs
#1123
opened Mar 8, 2024 by
MaierOli2010
Loading…
Update documentation in quantize.py
technique:qat
Regarding tfmot.quantization.keras (for quantization-aware training) APIs and docs
#1110
opened Jan 28, 2024 by
TheCrazyT
Loading…
Rewire TensorFlow to rely on tf_keras target.
technique:pruning
Regarding tfmot.sparsity.keras APIs and docs
#1097
opened Sep 14, 2023 by
copybara-service
Bot
Loading…
remove early out in xnnpack pruning policy model traversal
technique:pruning
Regarding tfmot.sparsity.keras APIs and docs
#1088
opened Aug 18, 2023 by
christian-steinmeyer
Loading…
add original weight name as prefix for mask and threshold weights
technique:pruning
Regarding tfmot.sparsity.keras APIs and docs
#1078
opened Jun 28, 2023 by
christian-steinmeyer
Loading…
allow registering custom layers in prune and clustering registries.
api-review
API review needed
technique:clustering
Regarding tfmot.clustering.keras APIs and docs
technique:pruning
Regarding tfmot.sparsity.keras APIs and docs
#1070
opened May 14, 2023 by
aaarrti
Loading…
Adds documentation for default_n_bit quantization scheme.
technique:qat
Regarding tfmot.quantization.keras (for quantization-aware training) APIs and docs
#1045
opened Feb 28, 2023 by
guynich
Loading…
Fix broken link in comprehensive_guide.ipynb
technique:pruning
Regarding tfmot.sparsity.keras APIs and docs
#1042
opened Feb 18, 2023 by
synandi
Loading…
Creates non-breaking changes where necessary in preparation for switching all of Keras to new serialization format.
technique:clustering
Regarding tfmot.clustering.keras APIs and docs
technique:pruning
Regarding tfmot.sparsity.keras APIs and docs
technique:qat
Regarding tfmot.quantization.keras (for quantization-aware training) APIs and docs
#1039
opened Feb 10, 2023 by
copybara-service
Bot
Loading…
Cast using FunctionType instead of input_signature
#1037
opened Feb 6, 2023 by
copybara-service
Bot
Loading…
Changes external references to Regarding tfmot.sparsity.keras APIs and docs
technique:qat
Regarding tfmot.quantization.keras (for quantization-aware training) APIs and docs
keras.utils.serialize_keras_object/deserialize_keras_object to legacy serialization API in preparation for switching all of Keras to new serialization format.
technique:pruning
#1036
opened Jan 19, 2023 by
copybara-service
Bot
Loading…
Update comprehensive_guide.ipynb
technique:pruning
Regarding tfmot.sparsity.keras APIs and docs
#1029
opened Nov 28, 2022 by
synandi
Loading…
Previous Next
ProTip!
Add no:assignee to see everything that’s not assigned.