Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Jan 16, 2025

• [BE] Add missing throw of `std::runtime_error` in scrc/cuda/utils.cpp (#144962)

  Show description for 13d35ea
  

  rec

  authored and

  pytorchmergebot

  committed
  13d35ea

  Copy full SHA for 13d35ea

• [export] Support module inputs for non strict mode. (#143925)

  Show description for 53256ed
  

  zhxchen17

  authored and

  pytorchmergebot

  committed
  53256ed

  Copy full SHA for 53256ed

• [BE] - Remove conda test and upload scripts and env variables from Workflows Part 1 (#144870)

  Show description for 519269a
  

  atalman

  authored and

  pytorchmergebot

  committed
  519269a

  Copy full SHA for 519269a

• Cholesky mps implementation (#144193)

  Show description for 727ae13
  

  2 peopleauthored and

  pytorchmergebot

  committed
  727ae13

  Copy full SHA for 727ae13

• [mps] Massage test_full_truncation to work only on the supported dtypes. (#144877)

  Show description for 1b34665
  

  dcci

  authored and

  pytorchmergebot

  committed
  1b34665

  Copy full SHA for 1b34665

• [aoti] Deduplicate "V.aot_compilation" and "V.graph.aot_mode" flags. [1/n] (#144709)

  Show description for 3d29de3
  

  zhxchen17

  authored and

  pytorchmergebot

  committed
  3d29de3

  Copy full SHA for 3d29de3

• Fix erroneous at_vreinterpretq_u16_bf16 call (#144883)

  Show description for 241a8a1
  

  swolchok

  authored and

  pytorchmergebot

  committed
  241a8a1

  Copy full SHA for 241a8a1

• Revert "Add flop formula for _scaled_mm (#144872)"

  Show description for 6559374
  pytorchmergebot

  committed
  6559374

  Copy full SHA for 6559374

• Update torch-xpu-ops commit pin (#144739)

  Show description for 6470b0e
  

  toyxu

  authored and

  pytorchmergebot

  committed
  6470b0e

  Copy full SHA for 6470b0e

• Add flop formula for _scaled_mm (#144872)

  Show description for f314522
  

  lw

  authored and

  pytorchmergebot

  committed
  f314522

  Copy full SHA for f314522

• Revert "Add tests for different dtypes with max autotune (#144721)"

  Show description for 1c29091
  pytorchmergebot

  committed
  1c29091

  Copy full SHA for 1c29091

• [inductor] fix index.Tensor fallback (#144736)

  Show description for 0c05832
  

  shunting314

  authored and

  pytorchmergebot

  committed
  0c05832

  Copy full SHA for 0c05832

• XFAIL test_save_load_checkpoint (#144927)

  Show description for 57d5659
  

  huydhn

  authored and

  pytorchmergebot

  committed
  57d5659

  Copy full SHA for 57d5659

• [Pipelining] Improve shape inference debug logging (#144929)

  Show description for 7d8c087
  

  wconstab

  authored and

  pytorchmergebot

  committed
  7d8c087

  Copy full SHA for 7d8c087

• restore rng generation for fbcode (#144819)

  Show description for 0b17c09
  

  ngimel

  authored and

  pytorchmergebot

  committed
  0b17c09

  Copy full SHA for 0b17c09

• Add strict kwarg to `nn.Module.set_submodule` and fix bug for non dot delineated strings (#143455)

  Show description for 49bdc41
  

  mariovas3

  authored and

  pytorchmergebot

  committed
  49bdc41

  Copy full SHA for 49bdc41

• [c10d][fr] Fix the bug when we still mark mismatch when there are match case (#144916)

  Show description for e3c4d1b
  

  fduwjj

  authored and

  pytorchmergebot

  committed
  e3c4d1b

  Copy full SHA for e3c4d1b

• Add tests for different dtypes with max autotune (#144721)

  Show description for 9e568cb
  

  exclamaforte

  authored and

  pytorchmergebot

  committed
  9e568cb

  Copy full SHA for 9e568cb

• OpenReg: Use device agnostic API (#144840)

  Show description for 52a6208
  

  Zhenbin-8

  authored and

  pytorchmergebot

  committed
  52a6208

  Copy full SHA for 52a6208

• [Quant][Inductor][X86] Separate binary post op fusion and lowering for qconv (#144318)

  Show description for 1230de4
  

  Xia-Weiwen

  authored and

  pytorchmergebot

  committed
  1230de4

  Copy full SHA for 1230de4

• Remove unnecessary once flag usage (#143255)

  Show description for 843627b
  

  cyyever

  authored and

  pytorchmergebot

  committed
  843627b

  Copy full SHA for 843627b

• [MPSInductor] Fix codegen regression (#144924)

  Show description for 41ec2e8
  

  malfet

  authored and

  pytorchmergebot

  committed
  41ec2e8

  Copy full SHA for 41ec2e8

• [MPSInductor] Properly convert index (#144917)

  Show description for 0550577
  

  malfet

  authored and

  pytorchmergebot

  committed
  0550577

  Copy full SHA for 0550577

• Revert "restore rng generation for fbcode (#144819)"

  Show description for d595b96
  pytorchmergebot

  committed
  d595b96

  Copy full SHA for d595b96

• symbolic_convert: Don't fail when we hit a undefined name (#144784)

  Show description for 6492851
  

  c00w

  authored and

  pytorchmergebot

  committed
  6492851

  Copy full SHA for 6492851

• Default Copies are not vectorized in v3.6.0 of cutlass (#144837)

  Show description for c8bcb22
  

  drisspg

  authored and

  pytorchmergebot

  committed
  c8bcb22

  Copy full SHA for c8bcb22

• speculation_log: Raise a unique error for divergence issues (#144785)

  Show description for 926f905
  

  c00w

  authored and

  pytorchmergebot

  committed
  926f905

  Copy full SHA for 926f905

• [inductor][BE] don't try/except ImportError for AttrsDescriptor versions (#144807)

  Show description for b90231a
  

  davidberard98

  authored and

  pytorchmergebot

  committed
  b90231a

  Copy full SHA for b90231a

• Apply Ruff fixes and pyupgrade to torch/jit (#144208)

  Show description for ee97d80
  

  cyyever

  authored and

  pytorchmergebot

  committed
  ee97d80

  Copy full SHA for ee97d80

• [export] handle buffer/input mutations for joint-graph (#144806)

  Show description for 774f21a
  

  pianpwk

  authored and

  pytorchmergebot

  committed
  774f21a

  Copy full SHA for 774f21a

• dynamic shape support for interpolate(antialias=True) backward (#141198)

  Show description for d7f45fc
  

  bdhirsh

  authored and

  pytorchmergebot

  committed
  d7f45fc

  Copy full SHA for d7f45fc

• support numbers as tensors for aten.copy(Tensor, Tensor) (#141161)

  Show description for 4831f89
  

  bdhirsh

  authored and

  pytorchmergebot

  committed
  4831f89

  Copy full SHA for 4831f89

Commits on Jan 15, 2025

• export AOTI_TORCH_EXPORT on Windows. (#140030)

  Show description for 2645fc4
  

  2 peopleauthored and

  pytorchmergebot

  committed
  2645fc4

  Copy full SHA for 2645fc4

• [ONNX] Use python_dispatcher in type promotion (#144801)

  Show description for fb4b5a9
  

  justinchuby

  authored and

  pytorchmergebot

  committed
  fb4b5a9

  Copy full SHA for fb4b5a9

• Enable s8s8s8 for qlinear with mkl-dnn (#139887)

  Show description for 7265dc0
  

  annop-w

  authored and

  pytorchmergebot

  committed
  7265dc0

  Copy full SHA for 7265dc0

Pagination

Previous

Next