Commits

Breadcrumbs

History for

onmaster

User selector

All users

Datepicker

All time

Commit History

Commits on Dec 21, 2018

• Update with clang format (pytorch#51)

  Show description for 4691d5b
  

  jianyuh

  authored and

  facebook-github-bot

  committed
  4691d5b

Commits on Dec 19, 2018

• Refactor to use FbgemmFP16 in packed gemm operator (pytorch#49)

  Show description for d5810be
  

  amylittleyang

  authored and

  facebook-github-bot

  committed
  d5810be

Commits on Dec 17, 2018

• add comments on col_offsets (pytorch#48)

  Show description for 1b3d970
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  1b3d970

Commits on Dec 5, 2018

• Final cleanup for avx2 isolation and consistent file names (pytorch#40)

  Show description for 3af8fe5
  

  dskhudia

  authored and

  facebook-github-bot

  committed
  3af8fe5

• avx2 intrinsic separation from OutputProcessing-inl.h (pytorch#38)

  Show description for 7e1b543
  

  dskhudia

  authored and

  facebook-github-bot

  committed
  7e1b543

• avx2 specific code in a separate file for QuantUtils (pytorch#29)

  Show description for bf398ec
  

  dskhudia

  authored and

  facebook-github-bot

  committed
  bf398ec

• Move avx2 specific code in different source files (pytorch#28)

  Show description for 114a461
  

  dskhudia

  authored and

  facebook-github-bot

  committed
  114a461

Commits on Dec 1, 2018

• Build fix with fbgemm shared lib (pytorch#31)

  Show description for 4fcc1f1
  

  dskhudia

  authored and

  facebook-github-bot

  committed
  4fcc1f1

Commits on Nov 30, 2018

• Only export symbols that are required while building shared library

  Show description for 4e9fa6c
  

  dskhudia

  authored and

  facebook-github-bot

  committed
  4e9fa6c

Commits on Nov 29, 2018

• sparse convolution output processing (pytorch#27)

  Show description for 027de07
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  027de07

Commits on Nov 27, 2018

• per-group and per-channel quantization (#14340)

  Show description for d4ee77f
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  d4ee77f

• fix group convention in B packing (pytorch#26)

  Show description for db52c82
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  db52c82

Commits on Nov 26, 2018

• remove unnecessary zero_point argument from constructors (#14323)

  Show description for ea47a69
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  ea47a69

• minimize code compiled with avx2 and header includes from them (#14313)

  Show description for a364b4f
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  a364b4f

Commits on Nov 22, 2018

• adding quantization utility functions (pytorch#19)

  Show description for 719734d
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  719734d

• optimize for symmetric quantization in requantization (pytorch#18)

  Show description for a1f11f8
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  a1f11f8

Commits on Nov 20, 2018

• Optimize parallelization performance (pytorch#15)

  Show description for 12c8362
  

  jianyuh

  authored and

  facebook-github-bot

  committed
  12c8362

• A function to check if we are running on a fbgemm supported cpu (pytorch#13)

  Show description for 3b7936e
  

  dskhudia

  authored and

  facebook-github-bot

  committed
  3b7936e

Commits on Nov 19, 2018

• clang-format (pytorch#11)

  Show description for 6d00b52
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  6d00b52

Commits on Nov 16, 2018

• grouped conv (pytorch#8)

  Show description for 25ab38b
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  25ab38b

• grouped (batched) gemm (pytorch#7)

  Show description for 8392eca
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  8392eca

Commits on Nov 13, 2018

• refactoring transpose code and comment

  Show description for 56d9537
  

  jspark1105

  authored and

  facebook-github-bot

  committed
  56d9537

Commits on Nov 11, 2018

• Fix the issue caused by isA(); Remove use of avx512 cast intrinsics for GCC 4.9.2 compatibility

  jianyuh

  committed
  1b56d08

Commits on Nov 8, 2018

• Sync with internal copy: Asymmetric padding; fbgemm2 -> fbgemm

  jianyuh

  committed
  428a0b6

Commits on Nov 6, 2018

• Add equals and metaEquals method to PackBMatrix

  jspark1105

  committed
  05e3a2f

Commits on Nov 5, 2018

• generalized conv_param_t and download third party libraries in build dir

  dskhudia

  committed
  b96bc0b

Commits on Nov 4, 2018

• Syncing with internal version. Fixes for Mac/clang build. Other minor fixes

  dskhudia

  committed
  690dbc2

Commits on Oct 30, 2018

• Initial commit

  dskhudia

  committed
  e85b5a1

Loading