pytorch/caffe2/perfkernels
Jongsoo Park a7fe200f5f [caffe2] simplify caffe2 code with fbgemm handling block size 1 emb (#33774)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/33774

Simplify caffe2 code using D19246900

Test Plan: CI

Reviewed By: jianyuh

Differential Revision: D20102410

fbshipit-source-id: 8de4d9cfac66898db0718ac6477339fd5e5428e3
2020-02-27 14:45:28 -08:00
..
__init__.py
adagrad.cc [caffe2] simplify caffe2 code with fbgemm handling block size 1 emb (#33774) 2020-02-27 14:45:28 -08:00
adagrad.h [caffe2] simplify caffe2 code with fbgemm handling block size 1 emb (#33774) 2020-02-27 14:45:28 -08:00
adagrad_avx.cc [caffe2] simplify caffe2 code with fbgemm handling block size 1 emb (#33774) 2020-02-27 14:45:28 -08:00
CMakeLists.txt
common.h
common_avx.cc
common_avx2.cc
common_avx512.cc
cvtsh_ss_bugfix.h
embedding_lookup.cc
embedding_lookup.h
embedding_lookup_avx2.cc
embedding_lookup_fused_8bit_rowwise_avx2.cc
embedding_lookup_fused_8bit_rowwise_idx_avx2.cc [pytorch][embeddingbag_8bit] Add include_last_offset option to Fused 8bit EmbeddingBag and parallelize the op (#32683) 2020-01-29 16:04:56 -08:00
embedding_lookup_idx.cc [pytorch][embeddingbag] Parallelize the EmbeddingBag operator (#4049) 2020-01-23 21:29:44 -08:00
embedding_lookup_idx.h [pytorch][embeddingbag] Parallelize the EmbeddingBag operator (#4049) 2020-01-23 21:29:44 -08:00
embedding_lookup_idx_avx2.cc [pytorch][embeddingbag] Parallelize the EmbeddingBag operator (#4049) 2020-01-23 21:29:44 -08:00
fused_8bit_rowwise_conversion.cc add conversion functions to embedding tables (#31083) 2020-01-08 16:56:12 -08:00
fused_8bit_rowwise_conversion.h add conversion functions to embedding tables (#31083) 2020-01-08 16:56:12 -08:00
fused_8bit_rowwise_conversion_avx2.cc optimize FloatToFused8BitRowwiseQuantized and Fused8BitRowwiseQuantizedToFloat (#31470) 2019-12-20 10:09:26 -08:00
fused_8bit_rowwise_embedding_lookup.cc
fused_8bit_rowwise_embedding_lookup.h
fused_8bit_rowwise_embedding_lookup_idx.cc [pytorch][embeddingbag_8bit] Add include_last_offset option to Fused 8bit EmbeddingBag and parallelize the op (#32683) 2020-01-29 16:04:56 -08:00
fused_8bit_rowwise_embedding_lookup_idx.h
hp_emblookup_codegen.py [pytorch][embeddingbag_8bit] Add include_last_offset option to Fused 8bit EmbeddingBag and parallelize the op (#32683) 2020-01-29 16:04:56 -08:00
math.h
math_cpu_avx2.cc
math_cpu_base.cc
typed_axpy.cc
typed_axpy.h
typed_axpy_avx.cc
typed_axpy_avx2.cc