Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Jul 25, 2022

• [reland] make ShardedTensor be a Tensor and nn.Parameter (#82089)

  Show description for 7ff121e
  

  wanchaol

  authored and

  pytorchmergebot

  committed
  7ff121e

  Copy full SHA for 7ff121e

• Re-enable iOS simulator tests (#82027)

  Show description for 40e99fe
  

  kit1980

  authored and

  pytorchmergebot

  committed
  40e99fe

  Copy full SHA for 40e99fe

• Fix capitalization (#82112)

  Show description for e7920c0
  

  ezyang

  authored and

  pytorchmergebot

  committed
  e7920c0

  Copy full SHA for e7920c0

• Pass stride overload, not overload packet; add aten.stride.default (#82083)

  Show description for 563f6c7
  

  ezyang

  authored and

  pytorchmergebot

  committed
  563f6c7

  Copy full SHA for 563f6c7

• All zero sparse tensors are always coalesced (#82085)

  Show description for c88fbae
  

  ezyang

  authored and

  pytorchmergebot

  committed
  c88fbae

  Copy full SHA for c88fbae

• [FSDP] Refactor casting of grad to full param dtype (#81574)

  Show description for 57a5662
  

  Andrew Gu

  authored and

  pytorchmergebot

  committed
  57a5662

  Copy full SHA for 57a5662

• [ONNX] Export aten::convolution (#81815)

  Show description for 1a9317c
  

  wschin

  authored and

  pytorchmergebot

  committed
  1a9317c

  Copy full SHA for 1a9317c

• script to monitor memory + cpu utilization (#82006)

  Show description for 6f2a88d
  

  clee2000

  authored and

  pytorchmergebot

  committed
  6f2a88d

  Copy full SHA for 6f2a88d

• Add some more SparseMeta support (#82086)

  Show description for 0e995f8
  

  ezyang

  authored and

  pytorchmergebot

  committed
  0e995f8

  Copy full SHA for 0e995f8

• [nn] Add support for +=, * and *= operations for nn.Sequential objects (#81279)

  Show description for 110cd72
  

  Ansh Radhakrishnan

  authored and

  pytorchmergebot

  committed
  110cd72

  Copy full SHA for 110cd72

• change rocblas.h -> rocblas/rocblas.h (#80849)

  Show description for 28262e1
  

  acxz

  authored and

  pytorchmergebot

  committed
  28262e1

  Copy full SHA for 28262e1

• [JIT] Add backwards compatibility test for old NonDeterminism ops list in ir.cpp (#82029)

  Show description for 7288ea4
  

  goldenxuett

  authored and

  pytorchmergebot

  committed
  7288ea4

  Copy full SHA for 7288ea4

• Change autogradNotImplementedFallback to utilize is_mutable and is_aliasing from FunctionSchema (#81917)

  Show description for 3c20b5b
  

  goldenxuett

  authored and

  pytorchmergebot

  committed
  3c20b5b

  Copy full SHA for 3c20b5b

• [JIT] Add is_aliasing method to FunctionSchema (#81916)

  Show description for eb2ea9a
  

  goldenxuett

  authored and

  pytorchmergebot

  committed
  eb2ea9a

  Copy full SHA for eb2ea9a

• [JIT] Fix AliasDB edge case for multi non existing alias sets in output (#81837)

  Show description for 5237677
  

  goldenxuett

  authored and

  pytorchmergebot

  committed
  5237677

  Copy full SHA for 5237677

• [JIT] Modify is_nondeterministic to utilize tags in schemaInfo and integrate with ir.cpp (#81836)

  Show description for fc3555c
  

  goldenxuett

  authored and

  pytorchmergebot

  committed
  fc3555c

  Copy full SHA for fc3555c

• s/Compiance/Compliance (#82087)

  Show description for 14968d5
  

  ezyang

  authored and

  pytorchmergebot

  committed
  14968d5

  Copy full SHA for 14968d5

• [composite compliance] put, take (#81094)

  Show description for db0e121
  

  kshitij12345

  authored and

  pytorchmergebot

  committed
  db0e121

  Copy full SHA for db0e121

• [ONNX] Fix ONNX aten::mul exporter with boolean inputs (#81671)

  Show description for d30784b
  

  wschin

  authored and

  pytorchmergebot

  committed
  d30784b

  Copy full SHA for d30784b

• Add Pull Request template for functorch (#81995)

  Show description for 767942d
  

  zou3519

  authored and

  pytorchmergebot

  committed
  767942d

  Copy full SHA for 767942d

• Turn on linting for functorch (#81987)

  Show description for f42ed3f
  

  zou3519

  authored and

  pytorchmergebot

  committed
  f42ed3f

  Copy full SHA for f42ed3f

• Delete some no-longer-neccessary debugging output (#81990)

  Show description for 5cb802a
  

  zou3519

  authored and

  pytorchmergebot

  committed
  5cb802a

  Copy full SHA for 5cb802a

• Add more functorch shards to PR CI (#82013)

  Show description for 68bd687
  

  zou3519

  authored and

  pytorchmergebot

  committed
  68bd687

  Copy full SHA for 68bd687

• Add ability to functorch tests via run_test.py (#82012)

  Show description for 5f4e8c0
  

  zou3519

  authored and

  pytorchmergebot

  committed
  5f4e8c0

  Copy full SHA for 5f4e8c0

• Add initial support for differentiable optimizers (#80938)

  Show description for 49b4f45
  

  Emilio Castillo

  authored and

  pytorchmergebot

  committed
  49b4f45

  Copy full SHA for 49b4f45

• Add test_make_fx_model_train example (#980) (#82011)

  Show description for 0894c49
  

  mostafaelhoushi

  authored and

  pytorchmergebot

  committed
  0894c49

  Copy full SHA for 0894c49

• Support non-standard bools in CUDA mode (#79393)

  Show description for 02550bc
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  02550bc

  Copy full SHA for 02550bc

• [composite compliance] matrix_exp (#81225)

  Show description for 5880a66
  

  kshitij12345

  authored and

  pytorchmergebot

  committed
  5880a66

  Copy full SHA for 5880a66

• [torchdynamo hash update] update the pinned torchdynamo hash (#82080)

  Show description for 83c6113
  pytorchmergebot

  committed
  83c6113

  Copy full SHA for 83c6113

Commits on Jul 24, 2022

• [ROCm]: Enable test_grad_layout_1devicemodule_1replicaperprocess (#82005)

  Show description for 594652f
  

  jaglinux

  authored and

  pytorchmergebot

  committed
  594652f

  Copy full SHA for 594652f

• Remove unused line (#82019)

  Show description for 64094d8
  

  wschin

  authored and

  pytorchmergebot

  committed
  64094d8

  Copy full SHA for 64094d8

• Do not suppress fallback exception in fake tensor (#82066)

  Show description for 5810391
  

  ezyang

  authored and

  pytorchmergebot

  committed
  5810391

  Copy full SHA for 5810391

• [torchdynamo hash update] update the pinned torchdynamo hash (#82069)

  Show description for 17730ac
  pytorchmergebot

  committed
  17730ac

  Copy full SHA for 17730ac

• [xla hash update] update the pinned xla hash (#81676)

  Show description for 51c3054
  pytorchmergebot

  committed
  51c3054

  Copy full SHA for 51c3054

• fix a warning uint to int changes a sign (#82030)

  Show description for 230f217
  

  Krovatkin

  authored and

  pytorchmergebot

  committed
  230f217

  Copy full SHA for 230f217

Pagination

Previous

Next