Pytorch_Quantize_impls icon indicating copy to clipboard operation
Pytorch_Quantize_impls copied to clipboard

Published 20 hours ago • verified publisher icon Enderdead
→

Metadata

Some recent Quantizing techniques on PyTorch

Results 3 Pytorch_Quantize_impls issues
Sort by recently updated
recently updated
newest added

Module eval decrease descent perf

Change weight on eval mod decrease Adam perf.

Enderdead avatar Enderdead

ShiftBatch2d efficiency

Fix ShiftBatch2d efficientness on ConvBinaryNet.

Enderdead avatar Enderdead

TODO

Capsule class

Create a capsule class in order to use Optuna lib for hyperparameter optimization.

Enderdead avatar Enderdead

TODO

← Metadata

72
Stars
10
Forks
Watchers

Owner

verified publisher icon Enderdead

Metadata

Some recent Quantizing techniques on PyTorch

Back