Commits

Breadcrumbs

History for

on9210e8f

User selector

All users

Datepicker

All time

Commit History

Commits on Mar 9, 2022

• [PyTorch] Move is_nested_tensor_impl & add get_nested_tensor_impl_or_null (pytorch#73928)

  Show description for b9c59dd
  

  swolchok

  authored and

  pytorchmergebot

  committed
  b9c59dd

  Copy full SHA for b9c59dd

• [pytorch] flatten_indices function should use vector::resize instead of reserve (pytorch#73831)

  Show description for ad6290f
  

  jiyuanzFB

  authored and

  pytorchmergebot

  committed
  ad6290f

  Copy full SHA for ad6290f

Commits on Mar 8, 2022

• [Quant] Qadd: Add qint8 support backed by xnnpack (pytorch#73663)

  Show description for 2e039b6
  

  digantdesai

  authored and

  pytorchmergebot

  committed
  2e039b6

  Copy full SHA for 2e039b6

• Use cub::DeviceSelect::UniqueByKey for EmbeddingBackward (pytorch#68376)

  Show description for ee8d7d8
  

  zasdfgbnm

  authored and

  pytorchmergebot

  committed
  ee8d7d8

  Copy full SHA for ee8d7d8

• Add Autocast support for Einsum (pytorch#71916)

  Show description for 1fbc08c
  

  Thiago Crepaldi

  authored and

  pytorchmergebot

  committed
  1fbc08c

  Copy full SHA for 1fbc08c

• Back out "Revert D34524207: [pytorch][PR] remove _s_where" (pytorch#73579)

  Show description for e47a5a6
  

  Natalia Gimelshein

  authored and

  pytorchmergebot

  committed
  e47a5a6

  Copy full SHA for e47a5a6

• [ROCM] Navi21 Enablement 5: Softmax kernels (pytorch#73545)

  Show description for 5dfbe52
  

  micmelesse

  authored and

  pytorchmergebot

  committed
  5dfbe52

  Copy full SHA for 5dfbe52

• record_function: add torchbind alternative API (pytorch#72301)

  Show description for 9ef5c67
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  9ef5c67

  Copy full SHA for 9ef5c67

Commits on Mar 7, 2022

• [RELAND] [cuDNN] Add a new optimized cuDNN RNN algorithm for small RNN hidden_size (pytorch#73211)

  Show description for 943085f
  

  mcarilli

  authored and

  pytorchmergebot

  committed
  943085f

  Copy full SHA for 943085f

Commits on Mar 6, 2022

• [Metal] Use MPSNNNeuronDescriptor for Mac Catalyst Target (pytorch#73718)

  Show description for 1d49711
  

  xta0

  authored and

  pytorchmergebot

  committed
  1d49711

  Copy full SHA for 1d49711

• [Metal] Remove unused shader code (pytorch#73636)

  Show description for e118b1d
  

  xta0

  authored and

  pytorchmergebot

  committed
  e118b1d

  Copy full SHA for e118b1d

Commits on Mar 5, 2022

• [XNNPACK] Expose available() to xnnpack namespace (pytorch#73617)

  Show description for 4b8106f
  

  digantdesai

  authored and

  pytorchmergebot

  committed
  4b8106f

  Copy full SHA for 4b8106f

• [Metal] Raise the minimum support version of iOS to 11.0 (pytorch#73635)

  Show description for 5dbec7c
  

  xta0

  authored and

  pytorchmergebot

  committed
  5dbec7c

  Copy full SHA for 5dbec7c

• CUDA Kernels: Use per-operator headers (1/4) (pytorch#71212)

  Show description for 723ba4e
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  723ba4e

  Copy full SHA for 723ba4e

• CPU Kernel: Use per-operator headers (pytorch#71137)

  Show description for 5816a95
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  5816a95

  Copy full SHA for 5816a95

Commits on Mar 4, 2022

• Revert D34554432: Back out "Revert D34524207: [pytorch][PR] remove _s_where"

  Show description for 5552563
  

  Natalia Gimelshein

  authored and

  pytorchmergebot

  committed
  5552563

  Copy full SHA for 5552563

• Back out "Revert D34524207: [pytorch][PR] remove _s_where" (pytorch#73579)

  Show description for 9c03c61
  

  Natalia Gimelshein

  authored and

  pytorchmergebot

  committed
  9c03c61

  Copy full SHA for 9c03c61

• [PyTorch] Make NestedTensor::dim() work (pytorch#73679)

  Show description for 88b19da
  

  swolchok

  authored and

  pytorchmergebot

  committed
  88b19da

  Copy full SHA for 88b19da

• Eliminate unused parameters in PyTorch (pytorch#73749)

  Show description for 72e4aab
  

  r-barnes

  authored and

  pytorchmergebot

  committed
  72e4aab

  Copy full SHA for 72e4aab

• Eliminate unused parameters in PyTorch (pytorch#73751)

  Show description for a7d2f39
  

  r-barnes

  authored and

  pytorchmergebot

  committed
  a7d2f39

  Copy full SHA for a7d2f39

Commits on Mar 3, 2022

• [Quant][core][devs] Removed support for non-quantized tensors in reflection_pad1d_cpu (pytorch#72485)

  Show description for a416022
  

  dzdang

  authored and

  pytorchmergebot

  committed
  a416022

  Copy full SHA for a416022

• Support 0s in out_size of FractionalMaxPoolNd

  Show description for 3186e36
  

  Emilio Castillo

  authored and

  pytorchmergebot

  committed
  3186e36

  Copy full SHA for 3186e36

Commits on Mar 2, 2022

• Support CSR to COO conversion in to_sparse(2). (pytorch#73642)

  Show description for 4168c87
  

  pearu

  authored and

  pytorchmergebot

  committed
  4168c87

  Copy full SHA for 4168c87

• [SR] Add new fb::split_and_squeeze op (pytorch#73252)

  Show description for 74fe570
  

  Mike Iovine

  authored and

  pytorchmergebot

  committed
  74fe570

  Copy full SHA for 74fe570

• Minimal NestedTensor (pytorch#72881)

  Show description for 484c0de
  

  cpuhrsch

  authored and

  pytorchmergebot

  committed
  484c0de

  Copy full SHA for 484c0de

• Eliminate unused parameters in PyTorch (pytorch#73625)

  Show description for 07df887
  

  r-barnes

  authored and

  pytorchmergebot

  committed
  07df887

  Copy full SHA for 07df887

• [cuDNN v8 API] cuDNN benchmark, convolution bwd / transposed convolution fwd, bfloat16, conv-bias-activation fusion (pytorch#60755)

  Show description for dab5e2a
  

  eqy

  authored and

  pytorchmergebot

  committed
  dab5e2a

  Copy full SHA for dab5e2a

• [ROCM] Navi21 Enablement 4: Normalization kernels (pytorch#73543)

  Show description for 7b8fc74
  

  micmelesse

  authored and

  pytorchmergebot

  committed
  7b8fc74

  Copy full SHA for 7b8fc74

Commits on Mar 1, 2022

• Write explicit meta-kernels for normal (pytorch#70089)

  Show description for eb0d370
  

  nkaretnikov

  authored and

  pytorchmergebot

  committed
  eb0d370

  Copy full SHA for eb0d370

• Unify checks for normal (pytorch#70087)

  Show description for dad0e0c
  

  nkaretnikov

  authored and

  pytorchmergebot

  committed
  dad0e0c

  Copy full SHA for dad0e0c

• Fix a typo: add a missing space (pytorch#70086)

  Show description for 08493e0
  

  nkaretnikov

  authored and

  pytorchmergebot

  committed
  08493e0

  Copy full SHA for 08493e0

• Bug fix: allow std 0 in the meta definition of normal_ (pytorch#70085)

  Show description for e9c1cce
  

  nkaretnikov

  authored and

  pytorchmergebot

  committed
  e9c1cce

  Copy full SHA for e9c1cce

• [Quant][core][gpu] Implemented support for relu in quantized conv operator using cudnn (pytorch#73337)

  Show description for 1957295
  

  dzdang

  authored and

  pytorchmergebot

  committed
  1957295

  Copy full SHA for 1957295

• [Quant][core][gpu] Implemented support for bias in quantized conv operator in cudnn (pytorch#73035)

  Show description for 4e54c1c
  

  dzdang

  authored and

  pytorchmergebot

  committed
  4e54c1c

  Copy full SHA for 4e54c1c

• [PyTorch] Add codegen unboxing ability (pytorch#69881)

  Show description for 9ce9803
  

  larryliu0820

  authored and

  pytorchmergebot

  committed
  9ce9803

  Copy full SHA for 9ce9803

Pagination

Previous

Next