pytorch/caffe2
Hector Yuen 27982d5711 fixes to layernorm emulation (#40422)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/40422

fix the remaining differences to the emulation of fp16 layernorm

Test Plan: unit test of layernorm

Reviewed By: venkatacrc

Differential Revision: D22182849

fbshipit-source-id: 8a45c21418517d65d7a41663d5ad2110d6b4677a
2020-06-23 11:51:13 -07:00
..
contrib fixes to layernorm emulation (#40422) 2020-06-23 11:51:13 -07:00
core Remove unecessary copy within blob serialization (#40096) 2020-06-16 14:45:59 -07:00
cuda_rtc
db
distributed
experiments
ideep
image
mobile
mpi
observers
onnx
operators Revert D22124313: Use Int8QuantParamsBlob to pass the scale and zeropoint params 2020-06-23 05:54:44 -07:00
opt rename the LayerNorm operator and add it to the replacement map (#40318) 2020-06-19 16:49:22 -07:00
perfkernels [caffe2] optimize 2/4-bit row-wise quantization (#387) 2020-06-19 21:28:31 -07:00
predictor
proto
python [caffe2] optimize 2/4-bit row-wise quantization (#387) 2020-06-19 21:28:31 -07:00
quantization Revert D22124313: Use Int8QuantParamsBlob to pass the scale and zeropoint params 2020-06-23 05:54:44 -07:00
queue
serialize Bumps readable file format version for torch.full inferring float from int values (#40089) 2020-06-16 15:09:40 -07:00
sgd Wrap Caffe2 (RowWise)SparseAdagrad fusion operator as a PT op (#39904) 2020-06-16 17:40:05 -07:00
share
test
transforms
utils Initial support for building on Ampere GPU, CUDA 11, cuDNN 8 (#39277) 2020-06-02 10:03:42 -07:00
video
.clang-format
__init__.py
c2_aten_srcs.bzl
CMakeLists.txt Set C++14 compatibility flag in torch_compile_options (#40399) 2020-06-23 07:10:22 -07:00
README.md
release-notes.md
requirements.txt
VERSION_NUMBER

Caffe2

Jenkins Build Status

Caffe2 is a lightweight, modular, and scalable deep learning framework. Building on the original Caffe, Caffe2 is designed with expression, speed, and modularity in mind.

Questions and Feedback

Please use Github issues (https://github.com/pytorch/pytorch/issues) to ask questions, report bugs, and request new features.

Further Resources on Caffe2.ai