Commits

Breadcrumbs

History for

on18dd5cd

User selector

All users

Datepicker

All time

Commit History

Commits on Dec 30, 2021

• [Operator Versioning][Test] Use hypothesis for better test input data and broader coverage (pytorch#70263)

  Show description for 18dd5cd
  

  cccclai

  authored and

  facebook-github-bot

  committed
  18dd5cd

  Copy full SHA for 18dd5cd

• [quant][fx][graphmode][be] Change the type for output of convert to be torch.nn.Module (pytorch#69959)

  Show description for c627211
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  c627211

  Copy full SHA for c627211

Commits on Dec 29, 2021

• Add testing across mem_formats to ModuleInfos (pytorch#69317)

  Show description for fb78a31
  

  mikaylagawarecki

  authored and

  facebook-github-bot

  committed
  fb78a31

  Copy full SHA for fb78a31

• Add Nondeterministic Tol to gradient test in test_modules (pytorch#69402)

  Show description for 14f4b91
  

  mikaylagawarecki

  authored and

  facebook-github-bot

  committed
  14f4b91

  Copy full SHA for 14f4b91

• Added antialias flag to interpolate (CPU only, bicubic) (pytorch#68819)

  Show description for d2abf3f
  

  vfdev-5

  authored and

  facebook-github-bot

  committed
  d2abf3f

  Copy full SHA for d2abf3f

• AdaptiveLogSoftmaxWithLoss no_batch_dim support (pytorch#69054)

  Show description for 8af39b7
  

  george-qi

  authored and

  facebook-github-bot

  committed
  8af39b7

  Copy full SHA for 8af39b7

• Remove backward op for slow dilated 3d convolution (pytorch#70068)

  Show description for fb736c7
  

  jbschlosser

  authored and

  facebook-github-bot

  committed
  fb736c7

  Copy full SHA for fb736c7

• [rnn,gru,lstm]cell : no batch dim (pytorch#70236)

  Show description for 2c67621
  

  kshitij12345

  authored and

  facebook-github-bot

  committed
  2c67621

  Copy full SHA for 2c67621

• [Operator Verioning][Edge] Update upgrader codegen with latest change (pytorch#70293)

  Show description for a0c99a8
  

  cccclai

  authored and

  facebook-github-bot

  committed
  a0c99a8

  Copy full SHA for a0c99a8

Commits on Dec 28, 2021

• quant tests: clean up logs about incorrect tensor copy (pytorch#70106)

  Show description for 00df885
  

  vkuzo

  authored and

  facebook-github-bot

  committed
  00df885

  Copy full SHA for 00df885

Commits on Dec 27, 2021

• Back out "Make TorchScript Preserve Fully Qualified Class Name for Python Exceptions"

  Show description for bf610f0
  

  Bo Wu

  authored and

  facebook-github-bot

  committed
  bf610f0

  Copy full SHA for bf610f0

Commits on Dec 25, 2021

• Add graph op replacement pass (pytorch#69915)

  Show description for 4ae71c8
  

  tugsbayasgalan

  authored and

  facebook-github-bot

  committed
  4ae71c8

  Copy full SHA for 4ae71c8

Commits on Dec 24, 2021

• Add utility methods to find an upgrader (pytorch#68355)

  Show description for df3cbcf
  

  tugsbayasgalan

  authored and

  facebook-github-bot

  committed
  df3cbcf

  Copy full SHA for df3cbcf

• Make TorchScript Preserve Fully Qualified Class Name for Python Exceptions (pytorch#70339)

  Show description for 911d527
  

  shunting314

  authored and

  facebook-github-bot

  committed
  911d527

  Copy full SHA for 911d527

• Revert D33234529: [NNC Testing] Randomized loop nest infrastructure

  Show description for 0ee663d
  

  Mikhail Zolotukhin

  authored and

  facebook-github-bot

  committed
  0ee663d

  Copy full SHA for 0ee663d

• Allow single node fusion for nvfuser (pytorch#70000)

  Show description for e429a68
  

  jjsjann123

  authored and

  facebook-github-bot

  committed
  e429a68

  Copy full SHA for e429a68

Commits on Dec 23, 2021

• Do not use ZeroTensor for inplace ops (pytorch#69998)

  Show description for 5ccf28d
  

  soulitzer

  authored and

  facebook-github-bot

  committed
  5ccf28d

  Copy full SHA for 5ccf28d

• Add forward AD formulas for {adaptive_,fractional_,}max_pool{2,3}d_{backward,} (pytorch#69884)

  Show description for 3116d87
  

  soulitzer

  authored and

  facebook-github-bot

  committed
  3116d87

  Copy full SHA for 3116d87

• [LTC] Upstream utils to extract BackendDevice from at::Tensor (pytorch#70069)

  Show description for ab57f6d
  

  Jiewen Tan

  authored and

  facebook-github-bot

  committed
  ab57f6d

  Copy full SHA for ab57f6d

• torch.linalg routines return torch.linalg.LinAlgError when a numerical error in the computation is found. (pytorch#68571)

  Show description for d100d98
  

  v0dro

  authored and

  facebook-github-bot

  committed
  d100d98

  Copy full SHA for d100d98

• Add the maximize flag to AdamW (pytorch#70146)

  Show description for a9c7d62
  

  Adnios

  authored and

  facebook-github-bot

  committed
  a9c7d62

  Copy full SHA for a9c7d62

• enable backward pass computation and communication overlap by prefetching all gather (pytorch#70235)

  Show description for b15212c
  

  zhaojuanmao

  authored and

  facebook-github-bot

  committed
  b15212c

  Copy full SHA for b15212c

• [NNC Testing] Randomized loop nest infrastructure (pytorch#70174)

  Show description for 1d09458
  

  anijain2305

  authored and

  facebook-github-bot

  committed
  1d09458

  Copy full SHA for 1d09458

• [quant][fx][graphmode] Add backend_config_dict for standalone module (pytorch#70150)

  Show description for 656d2a7
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  656d2a7

  Copy full SHA for 656d2a7

• Revert D33172665: [LTC] Upstream utils to extract BackendDevice from at::Tensor

  Show description for 795af15
  

  suo

  authored and

  facebook-github-bot

  committed
  795af15

  Copy full SHA for 795af15

• [LTC] Upstream utils to extract BackendDevice from at::Tensor (pytorch#70069)

  Show description for 121d067
  

  Jiewen Tan

  authored and

  facebook-github-bot

  committed
  121d067

  Copy full SHA for 121d067

Commits on Dec 22, 2021

• Remove backward ops for NNPACK spatial convolution (pytorch#70305)

  Show description for 2e94a0d
  

  jbschlosser

  authored and

  facebook-github-bot

  committed
  2e94a0d

  Copy full SHA for 2e94a0d

• TestMathBits: test with neg and conj bit set (pytorch#68948)

  Show description for 7cdfd86
  

  peterbell10

  authored and

  facebook-github-bot

  committed
  7cdfd86

  Copy full SHA for 7cdfd86

• [C++ API] Added missing nearest-exact mode and anti-alias flag (pytorch#69318)

  Show description for ce9a2f8
  

  vfdev-5

  authored and

  facebook-github-bot

  committed
  ce9a2f8

  Copy full SHA for ce9a2f8

• make inverse call linalg_inv (pytorch#70276)

  Show description for 56969bf
  

  Natalia Gimelshein

  authored and

  facebook-github-bot

  committed
  56969bf

  Copy full SHA for 56969bf

• Install TensorRT lib on oss docker and enable fx2trt unit test (pytorch#70203)

  Show description for 31c7e5d
  

  wushirong

  authored and

  facebook-github-bot

  committed
  31c7e5d

  Copy full SHA for 31c7e5d

• [LTC] Upstream LTCTensorImpl (pytorch#70062)

  Show description for e02d836
  

  Jiewen Tan

  authored and

  facebook-github-bot

  committed
  e02d836

  Copy full SHA for e02d836

• [nnc] Add a run method to TensorExprKernel that takes in output tensors (pytorch#69477)

  Show description for 4dec15e
  

  navahgar

  authored and

  facebook-github-bot

  committed
  4dec15e

  Copy full SHA for 4dec15e

• Back out "[Quant] Added 4 bit support for embedding quantized module" (pytorch#70273)

  Show description for b613fbd
  

  digantdesai

  authored and

  facebook-github-bot

  committed
  b613fbd

  Copy full SHA for b613fbd

• Back out "[Quant][Eager] Added 4 bit support for eager mode quantization flow" (pytorch#70272)

  Show description for 47ba28f
  

  digantdesai

  authored and

  facebook-github-bot

  committed
  47ba28f

  Copy full SHA for 47ba28f

Pagination

Previous

Next