frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

fx quant: enable linear-bn1d fusion for PTQ (#66484)

Browse Source 
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/66484

https://github.com/pytorch/pytorch/pull/50748 added linear - bn1d fusion
in Eager mode, for PTQ only. This PR also enables this in FX graph mode.

We reuse the existing conv-bn-relu fusion handler, renaming `conv` to
`conv_or_linear` for readability.

The QAT version is saved for a future PR, for both eager and FX graph.

Test Plan:
```
python test/test_quantization.py TestFuseFx.test_fuse_linear_bn_eval
```

Imported from OSS

Reviewed By: bdhirsh

Differential Revision: D31575392

fbshipit-source-id: f69d80ef37c98cbc070099170e335e250bcdf913
This commit is contained in:
Vasiliy Kuznetsov
2021-10-18 10:04:47 -07:00
committed by Facebook GitHub Bot
parent 9d287d0b63
commit d549c8de78
5 changed files with 42 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
torch/quantization/fx/fusion_patterns.py
View File

@ -8,6 +8,6 @@ here.
"""
from torch.ao.quantization.fx.fusion_patterns import (
FuseHandler,
ConvBNReLUFusion,
ConvOrLinearBNReLUFusion,
ModuleReLUFusion
)