pytorch/torch/nn/quantized
Digant Desai b613fbdbf2 Back out "[Quant] Added 4 bit support for embedding quantized module" (#70273)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/70273

Original commit changeset: 73e63383cf60

Original Phabricator Diff: D33152674 (9f512e129b)

Test Plan: CI

Reviewed By: larryliu0820

Differential Revision: D33268459

fbshipit-source-id: 051bfcbbad3fa083301a3cea508d00946d6db881
2021-12-21 21:28:04 -08:00
..
_reference [Quant][fx] Lower reference conv[1-3]d module (#69228) 2021-12-14 11:23:39 -08:00
dynamic Adding Dynamic Conv and ConvT ops/modules (#68176) 2021-11-15 16:42:25 -08:00
modules Back out "[Quant] Added 4 bit support for embedding quantized module" (#70273) 2021-12-21 21:28:04 -08:00
__init__.py
functional.py add a quantized hardsigmoid inplace variant (#65740) 2021-11-03 07:35:31 -07:00