Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Oct 6, 2023

• feat(optim): Add adadelta multi_tensor support for complex, with `has_complex` shortcut (#110631)

  Show description for 57e9969
  

  jon-chuang

  authored and

  pytorchmergebot

  committed
  57e9969

  Copy full SHA for 57e9969

• feat(optim): Add `NAdam`support for complex, with `has_complex` shortcut (#110634)

  Show description for 11047be
  

  jon-chuang

  authored and

  pytorchmergebot

  committed
  11047be

  Copy full SHA for 11047be

• feat(optim): Add `RAdam` support for complex, with `has_complex` shortcut (#110635)

  Show description for 347ea3f
  

  jon-chuang

  authored and

  pytorchmergebot

  committed
  347ea3f

  Copy full SHA for 347ea3f

• [export] Update ArgumentSpec definition. (#110612)

  Show description for be5dc3a
  

  zhxchen17

  authored and

  pytorchmergebot

  committed
  be5dc3a

  Copy full SHA for be5dc3a

• [aotinductor] Fix benchmarks with self.autocast (#110490)

  Show description for 83061ee
  

  angelayi

  authored and

  pytorchmergebot

  committed
  83061ee

  Copy full SHA for 83061ee

• [ez] Remove print in heuristics aggregation (#110621)

  Show description for 8a09fe4
  

  clee2000

  authored and

  pytorchmergebot

  committed
  8a09fe4

  Copy full SHA for 8a09fe4

• Revert "Multiprocessing support for NT (#110292)"

  Show description for dac895c
  pytorchmergebot

  committed
  dac895c

  Copy full SHA for dac895c

• Added a UserWarning when using torch.{std,var,std_mean,std_var} with dof<=0 (#109824)

  Show description for 555c83d
  

  tringwald

  authored and

  pytorchmergebot

  committed
  555c83d

  Copy full SHA for 555c83d

• Revert "pin_memory support for NT (#110404)"

  Show description for 81ce5d5
  pytorchmergebot

  committed
  81ce5d5

  Copy full SHA for 81ce5d5

• [Reland2] Remove calls of c10::either (#110487)

  Show description for 11b3210
  

  cyyever

  authored and

  pytorchmergebot

  committed
  11b3210

  Copy full SHA for 11b3210

Commits on Oct 5, 2023

• Revert "Update singleton int to error when inequality relation is undefined (#110044)"

  Show description for 330db82
  pytorchmergebot

  committed
  330db82

  Copy full SHA for 330db82

• Revert "Support SingletonSymNode mul with coefficient (#110369)"

  Show description for 1c3fae4
  pytorchmergebot

  committed
  1c3fae4

  Copy full SHA for 1c3fae4

• Revert "Update custom Function preserve torch function when inputs returned as-is (#109825)"

  Show description for 236afe7
  pytorchmergebot

  committed
  236afe7

  Copy full SHA for 236afe7

• Revert "Add symbolic singleton int (#110370)"

  Show description for fdf6055
  pytorchmergebot

  committed
  fdf6055

  Copy full SHA for fdf6055

• Revert "Symintify guards.cpp (#110371)"

  Show description for 585e2bd
  pytorchmergebot

  committed
  585e2bd

  Copy full SHA for 585e2bd

• Revert "Use is_symbolic instead of testing isinstance in some place (#110372)"

  Show description for bcd44da
  pytorchmergebot

  committed
  bcd44da

  Copy full SHA for bcd44da

• Replace enforce_dtype with dtype in ShardedTensor.gather (#110561)

  Show description for 5d96347
  

  yingufan

  authored and

  pytorchmergebot

  committed
  5d96347

  Copy full SHA for 5d96347

• Add functional collective all_to_all_single and support it in Inductor (#110195)

  Show description for f274c7b
  

  ezyang

  authored and

  pytorchmergebot

  committed
  f274c7b

  Copy full SHA for f274c7b

• perf(inductor): use for loop with shortcut in `Optimizer`s to speedup against list comprehensions (e.g. complex conversion) (#110613)

  Show description for df7d01a
  

  jon-chuang

  authored and

  pytorchmergebot

  committed
  df7d01a

  Copy full SHA for df7d01a

• [quant][pt2e] Refactor conv related annotation for XNNPACKQuantizer (#110308)

  Show description for 7b60421
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  7b60421

  Copy full SHA for 7b60421

• [BE] Get rid of code duplication (#110619)

  Show description for be02103
  

  malfet

  authored and

  pytorchmergebot

  committed
  be02103

  Copy full SHA for be02103

• [BE] Use nested namespaces in autograd/templates (#110618)

  Show description for 82e353f
  

  malfet

  authored and

  pytorchmergebot

  committed
  82e353f

  Copy full SHA for 82e353f

• Set _diffThreshold on our TestCase (#110603)

  Show description for cae5371
  

  albanD

  authored and

  pytorchmergebot

  committed
  cae5371

  Copy full SHA for cae5371

• [BE]: Enable some basic pytest style rules (#110362)

  Show description for 668eb55
  

  Skylion007

  authored and

  pytorchmergebot

  committed
  668eb55

  Copy full SHA for 668eb55

• [dtensor] add grad placements kwarg to to_local API (#110629)

  Show description for c95cf4b
  

  wanchaol

  authored and

  pytorchmergebot

  committed
  c95cf4b

  Copy full SHA for c95cf4b

• Add option to flop counter formula registration to get raw values (#110591)

  Show description for ada6550
  

  Chillee

  authored and

  pytorchmergebot

  committed
  ada6550

  Copy full SHA for ada6550

• [torch] easy missing move in aoti_runtime/model.h (#110469)

  Show description for 9e72c9c
  

  swolchok

  authored and

  pytorchmergebot

  committed
  9e72c9c

  Copy full SHA for 9e72c9c

• [dynamo, logging] Report name of defining class along side function name in Dynamo logs (#110190)

  Show description for 71beca4
  

  williamwen42

  authored and

  pytorchmergebot

  committed
  71beca4

  Copy full SHA for 71beca4

• fix(optim): adagrad sparse multitensor incorrect early exit (#110454)

  Show description for c99de9f
  

  jon-chuang

  authored and

  pytorchmergebot

  committed
  c99de9f

  Copy full SHA for c99de9f

• Back out "[Inductor] Break the loop fusion when node2 depends on node1 mutations (#109172)" (#110622)

  Show description for ecdd1bc
  

  ckluk2

  authored and

  pytorchmergebot

  committed
  ecdd1bc

  Copy full SHA for ecdd1bc

• [state_dict][1/N] Implement the basic functions of distributed.checkpoint._state_dict (#105902)

  Show description for 8861634
  

  fegin

  authored and

  pytorchmergebot

  committed
  8861634

  Copy full SHA for 8861634

• [aotinductor] Avoid generating redundant kernel loading code (#110510)

  Show description for 298f01d
  

  desertfire

  authored and

  pytorchmergebot

  committed
  298f01d

  Copy full SHA for 298f01d

• [AOTInductor] ProxyExecutor support Dynamic Shape (#110526)

  Show description for f1b9446
  

  SherlockNoMad

  authored and

  pytorchmergebot

  committed
  f1b9446

  Copy full SHA for f1b9446

• Add 9.0a to cpp_extension supported compute archs (#110587)

  Show description for a0cea51
  

  dzhulgakov

  authored and

  pytorchmergebot

  committed
  a0cea51

  Copy full SHA for a0cea51

• Bump pillow from 9.5.0 to 10.0.1 in /.ci/docker (#110494)

  Show description for c89d35a
  dependabot[bot]

  authored
  c89d35a

  Copy full SHA for c89d35a

Pagination

Previous

Next