tensorflow/model-optimization
A toolkit to optimize ML models for deployment for Keras and TensorFlow, including quantization and pruning.
PythonApache-2.0
Pinned issues
Issues
- 1
- 5
- 1
Model Pruning with Yolo Object Detection Model
#1018 opened - 2
- 5
- 1
Input and resource quantization
#1003 opened - 7
- 3
- 2
about the Quantize layer when trans model
#997 opened - 2
QAT for subclass inside the subclass
#996 opened - 0
1
#995 opened - 2
Allow quantization of tied weights
#994 opened - 5
- 0
Cannot joblib serialize pruned models
#990 opened - 2
- 8
- 8
- 9
- 1
- 4
- 13
Full Int8 QAT not working
#974 opened - 4
Pruning only works for small batch sizes
#973 opened - 1
- 4
- 7
- 15
QAT model saving bug: Unable to save function b'__inference_separable_conv2d_layer_call_fn_961'
#964 opened - 1
Stripping Quantized Model
#958 opened - 0
- 0
Quantize naive !!!
#956 opened - 0
- 0
Unable to quantize to 4-bits
#950 opened - 0
- 0
QAT support for LayerNormalization
#942 opened - 1
Unsupported operations when applying tfmot
#941 opened - 3
Not able to cluster Conv1DTranspose layer
#940 opened - 1
[clustering] Possible wrong call of the centroids initializer from the ClusterWeights wrapper
#939 opened - 0
- 3
- 1
Support for tensorflow hub layers
#924 opened - 0
GELU - Keras Layer
#922 opened - 1
- 5
- 1
Learnable Scale Quantizer?
#903 opened - 1
- 1
- 4
Model input names are not preserved during TFLite conversion when inference_input_type is tf.int8
#889 opened - 2
Pruning without training
#888 opened - 3
- 3
How to use default n bit in QAT ?
#885 opened - 7