Commits

Breadcrumbs

History for

on4451eb2

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 30, 2022

• Add layout and blocksize arguments to Tensor.to_sparse method (pytorch#89502)

  Show description for 76c6dfe
  

  pearu

  authored and

  pytorchmergebot

  committed
  76c6dfe

  Copy full SHA for 76c6dfe

• [MPS] Enable fp16 for linear backward (pytorch#89774)

  Show description for f2308b1
  

  kulinseth

  authored and

  pytorchmergebot

  committed
  f2308b1

  Copy full SHA for f2308b1

• [jiterator, complex32] lerp : cuda (pytorch#75584)

  Show description for b5ad909
  

  khushi-411

  authored and

  pytorchmergebot

  committed
  b5ad909

  Copy full SHA for b5ad909

• Revert "Call _sdp_attention in nn.functional.mha (pytorch#89470)"

  Show description for f1415b8
  pytorchmergebot

  committed
  f1415b8

  Copy full SHA for f1415b8

• Revert "Add bits tensor types (pytorch#88594)"

  Show description for 4cc5be3
  pytorchmergebot

  committed
  4cc5be3

  Copy full SHA for 4cc5be3

• Row and column select support for block compressed sparse tensors (pytorch#88733)

  Show description for 296e1ba
  

  pearu

  authored and

  pytorchmergebot

  committed
  296e1ba

  Copy full SHA for 296e1ba

• fix RowwiseMoments vectorization issue on CPU (pytorch#84404)

  Show description for 87d18cf
  

  mingfeima

  authored and

  pytorchmergebot

  committed
  87d18cf

  Copy full SHA for 87d18cf

Commits on Nov 29, 2022

• Guard the boundary of index computed in compute_source_index_and_lambda (pytorch#89252)

  Show description for 620994c
  

  Jiong Gong

  authored and

  pytorchmergebot

  committed
  620994c

  Copy full SHA for 620994c

• Refactoring to share vectorization code for int8/uint8. (pytorch#89650)

  Show description for a78467f
  

  alexsamardzic

  authored and

  pytorchmergebot

  committed
  a78467f

  Copy full SHA for a78467f

• Call _sdp_attention in nn.functional.mha (pytorch#89470)

  Show description for 4d7ec30
  

  drisspg

  authored and

  pytorchmergebot

  committed
  4d7ec30

  Copy full SHA for 4d7ec30

Commits on Nov 28, 2022

• Add bits tensor types (pytorch#88594)

  Show description for f3b1315
  

  angelayi

  authored and

  pytorchmergebot

  committed
  f3b1315

  Copy full SHA for f3b1315

• [CUDA graphs] Add warning if captured graph is empty (pytorch#88754)

  Show description for 0057be3
  

  Aidyn-A

  authored and

  pytorchmergebot

  committed
  0057be3

  Copy full SHA for 0057be3

• [Vulkan][TCC] Fix conv2d pack biases (pytorch#89568)

  Show description for 5abe454
  

  manuelcandales

  authored and

  pytorchmergebot

  committed
  5abe454

  Copy full SHA for 5abe454

• Add meta implementation for _efficientzerotensor (pytorch#88936)

  Show description for 2e0cd7c
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  2e0cd7c

  Copy full SHA for 2e0cd7c

• [ROCm] Replace layer_norm_grad_input_kernel with cuComputeGradInput for ROCm (pytorch#87726)

  Show description for cf4969d
  

  hubertlu-tw

  authored and

  pytorchmergebot

  committed
  cf4969d

  Copy full SHA for cf4969d

Commits on Nov 25, 2022

• TorchDynamo: weight prepack for mkl linear (pytorch#89109)

  Show description for 0c4f3db
  

  XiaobingSuper

  authored and

  pytorchmergebot

  committed
  0c4f3db

  Copy full SHA for 0c4f3db

• TorchDynamo: weight prepack for onednn convolution external call (pytorch#88988)

  Show description for 07151a6
  

  XiaobingSuper

  authored and

  pytorchmergebot

  committed
  07151a6

  Copy full SHA for 07151a6

Commits on Nov 24, 2022

• Added vectorized CPU code for uint8_t datatype. (pytorch#89284)

  Show description for 83666f1
  

  alexsamardzic

  authored and

  pytorchmergebot

  committed
  83666f1

  Copy full SHA for 83666f1

• [QAT] Check the value of numel to avoid segfault (pytorch#81547)

  Show description for b515c1d
  

  hguandl

  authored and

  pytorchmergebot

  committed
  b515c1d

  Copy full SHA for b515c1d

• Symintify embedding (pytorch#89327)

  Show description for 4cb6bbb
  

  nkaretnikov

  authored and

  pytorchmergebot

  committed
  4cb6bbb

  Copy full SHA for 4cb6bbb

• Symintified layer_norm (pytorch#89466)

  Show description for f0e5bc4
  

  tissue3

  authored and

  pytorchmergebot

  committed
  f0e5bc4

  Copy full SHA for f0e5bc4

• Install missing VSX headers (POWER) (pytorch#85547)

  Show description for fdb2dd1
  

  Flamefire

  authored and

  pytorchmergebot

  committed
  fdb2dd1

  Copy full SHA for fdb2dd1

• Add c10:: namespace in front of optional (pytorch#89605)

  Show description for 0bde514
  

  malfet

  authored and

  pytorchmergebot

  committed
  0bde514

  Copy full SHA for 0bde514

Commits on Nov 23, 2022

• [primTorch] Enable regex error testing for some refs (pytorch#87765)

  Show description for 0a1a530
  

  nkaretnikov

  authored and

  pytorchmergebot

  committed
  0a1a530

  Copy full SHA for 0a1a530

• Rectify native_batch_norm schema by splitting it into two legit schemas (pytorch#88697)

  Show description for 8695f0c
  

  janeyx99

  authored and

  pytorchmergebot

  committed
  8695f0c

  Copy full SHA for 8695f0c

• FFT: disable dimension wrapping for scalar tensors (pytorch#89234)

  Show description for ac19c5b
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  ac19c5b

  Copy full SHA for ac19c5b

• [quant][be] Refactor the error checking code for quantize_per_channel op (pytorch#89271)

  Show description for 128faf2
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  128faf2

  Copy full SHA for 128faf2

• Call symint::sizes() instead of sizes() on convolution error messages. (pytorch#89549)

  Show description for 795473f
  

  ysiraichi

  authored and

  pytorchmergebot

  committed
  795473f

  Copy full SHA for 795473f

• Fix Use-after-Free in qembeddingbag_byte_prepack_out (pytorch#84750)

  Show description for 7594e04
  

  Flamefire

  authored and

  pytorchmergebot

  committed
  7594e04

  Copy full SHA for 7594e04

• Symintify select (pytorch#89326)

  Show description for 07dd2fe
  

  nkaretnikov

  authored and

  pytorchmergebot

  committed
  07dd2fe

  Copy full SHA for 07dd2fe

• Added implementation and tests for MPS Hardswish (pytorch#87952)

  Show description for 4935b59
  

  thomaslin2020

  authored and

  pytorchmergebot

  committed
  4935b59

  Copy full SHA for 4935b59

Commits on Nov 22, 2022

• [Vulkan][TCC] Add tests for quantize_per_tensor and dequantize (pytorch#89496)

  Show description for ea920a1
  

  manuelcandales

  authored and

  pytorchmergebot

  committed
  ea920a1

  Copy full SHA for ea920a1

• [ROCm] Optimize layer norm backward kernel for ROCm (pytorch#87635)

  Show description for 74e62a1
  

  hubertlu-tw

  authored and

  pytorchmergebot

  committed
  74e62a1

  Copy full SHA for 74e62a1

• Fix vectorized trigonometric functions for VSX (pytorch#86453)

  Show description for 177baf3
  

  Flamefire

  authored and

  pytorchmergebot

  committed
  177baf3

  Copy full SHA for 177baf3

• Make aten.copy preserve strides (hf_Longformer) (pytorch#89464)

  Show description for d9cbe77
  

  ezyang

  authored and

  pytorchmergebot

  committed
  d9cbe77

  Copy full SHA for d9cbe77

Pagination

Previous

Next