Commits

Commits on Sep 16, 2024

Remove unused weight_modifiable_node_names property (#2965 )
andrey-churkin
committedSep 16, 2024

Commits on Sep 12, 2024

Type mismatch in onnx quantize_with_accuracy_control validation_fn (#2963 )
andrey-churkin
committedSep 12, 2024

Commits on Sep 10, 2024

Fix segmentation fault (#2955 )
andrey-churkin
committedSep 10, 2024

Commits on Sep 5, 2024

Extend NNCF logging by the device information where quantization was done (#2931 )
andrey-churkin
committedSep 5, 2024

Commits on Aug 29, 2024

Preserve the constant subgraph (#2902 )
andrey-churkin
committedAug 29, 2024

Commits on Aug 22, 2024

Replace the exception with a warning (#2901 )
andrey-churkin
committedAug 22, 2024

Commits on Jul 19, 2024

Revert changes (#2827 )
andrey-churkin
committedJul 19, 2024

Commits on Jul 18, 2024

Use a graph to find the output port id for the activation quantization target point (#2802 )
andrey-churkin
committedJul 18, 2024

Commits on Jul 12, 2024

Fix test (#2806 )
andrey-churkin
committedJul 12, 2024

Commits on Jul 10, 2024

Remove unused code (#2789 )
andrey-churkin
committedJul 10, 2024

Commits on Jul 1, 2024

Remove reference data for TF 2.8 from the tests (#2770 )
andrey-churkin
committedJul 1, 2024
Support TF 2.15 (#2609 )
andrey-churkin
committedJul 1, 2024

Commits on Jun 10, 2024

Update calibrate tool (#2721 )
andrey-churkin
committedJun 10, 2024

Commits on Mar 26, 2024

Support string dtype for OV backend (#2602 )
andrey-churkin
committedMar 26, 2024

Commits on Mar 8, 2024

Fix anomaly_stfpm_quantize_with_accuracy_control for Win (#2550 )
andrey-churkin
committedMar 8, 2024
Introduce an ignored subgraph (#2548 )
andrey-churkin
committedMar 8, 2024

Commits on Mar 1, 2024

Fix remove_shapeof_subgraphs (#2536 )
andrey-churkin
committedMar 1, 2024

Commits on Feb 22, 2024

Add quantize_with_accuracy_control example for ONNX (#2465 )
andrey-churkin
committedFeb 22, 2024

Commits on Feb 16, 2024

Fix TypeError: revert_operations_to_floating_point_precision() missing 2 required positional arguments (#2484 )
andrey-churkin
committedFeb 16, 2024

Commits on Feb 12, 2024

Add the AAQ algorithm for the ONNX backend (#2399 )
andrey-churkin
committedFeb 12, 2024

Commits on Feb 8, 2024

Add progress bar for calculating ranking scores (#2459 )
andrey-churkin
committedFeb 8, 2024

Commits on Feb 6, 2024

nncf.quantize_with_accuracy_control fails with tune_hyperparams=True (#2450 )
andrey-churkin
committedFeb 6, 2024

Commits on Jan 18, 2024

Remove unused ONNXNullBiasInsertionCommand command (#2403 )
andrey-churkin
committedJan 18, 2024

Commits on Jan 9, 2024

Fix TypeError: Object of type RestoreMode is not JSON serializable (#2378 )
andrey-churkin
committedJan 9, 2024

Commits on Dec 22, 2023

Add backup precision option (#2299 )
andrey-churkin
committedDec 22, 2023

Commits on Nov 23, 2023

OVWeightUpdateCommand copies a constant (#2277 )

andrey-churkin
and
KodiaqQ
committedNov 23, 2023

Commits on Oct 26, 2023

Fix index out of range (#2224 )
andrey-churkin
committedOct 26, 2023

Commits on Oct 25, 2023

Enable SmoothQuant only for OV backend (#2208 )
andrey-churkin
committedOct 25, 2023

Commits on Oct 17, 2023

Add SmoothQuant and ChannelAlignment to HyperparameterTuner (#2154 )
andrey-churkin
committedOct 17, 2023

Commits on Oct 16, 2023

Optimize NNCFGraph.get_node_by_name() method (#2190 )
andrey-churkin
committedOct 16, 2023
Fix multithreading (#2198 )

andrey-churkin
and
alexsu52
committedOct 16, 2023

Commits on Oct 12, 2023

Update reference (#2182 )
andrey-churkin
committedOct 12, 2023

Commits on Oct 6, 2023

Update calibrate tool (#2174 )
andrey-churkin
committedOct 6, 2023

Commits on Sep 12, 2023

Add support for the dump_intermediate_model parameter (#2086 )
andrey-churkin
committedSep 12, 2023

Commits on Sep 4, 2023

Reworking the placement of metatype groups (#2094 )
andrey-churkin
committedSep 4, 2023