Commits

Branch selector

 main

User selector

ydwu4

Datepicker

All time

Commit History

Commits on Nov 5, 2024

• [hop free symbols] refactor tensor.to_list implementation to call wrap_fx_proxy. (#139663)

  Show description for 6734cb7
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  6734cb7

  Copy full SHA for 6734cb7

Commits on Nov 4, 2024

• [hop free symbols][refactor] lift freevar to parent graph before lifting to subgraph (#138559)

  Show description for 397938b
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  397938b

  Copy full SHA for 397938b

• [hop free symbols] replace ctx.save_for_backward to support symints/ints (#138737)

  Show description for c5b7969
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  c5b7969

  Copy full SHA for c5b7969

• [hop free symbols][refactor] make map's save_for_backward to handle int (#138558)

  Show description for ac20d0f
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  ac20d0f

  Copy full SHA for ac20d0f

• [hop free symbols][refactor] make create_graph_input always take example_value (#138428)

  Show description for dc3a6a9
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  dc3a6a9

  Copy full SHA for dc3a6a9

• [hop free symbols][refactor] make bound_symbols a dictionary (#138345)

  Show description for 54c69a7
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  54c69a7

  Copy full SHA for 54c69a7

Commits on Oct 31, 2024

• [cond] make cond not throw warnings on constant pred in eager mode (#138837)

  Show description for 3842993
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  3842993

  Copy full SHA for 3842993

Commits on Oct 26, 2024

• [hop] always trace subgraph with fake to support .item in eager mode (#138771)

  Show description for 0ac9a66
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  0ac9a66

  Copy full SHA for 0ac9a66

Commits on Oct 25, 2024

• [scan] better error handling and remove redundant tests (#137967)

  Show description for c6bb9b5
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  c6bb9b5

  Copy full SHA for c6bb9b5

Commits on Oct 17, 2024

• [cond] support lifted symint inputs in subgraph (#137519)

  Show description for 3087b5e
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  3087b5e

  Copy full SHA for 3087b5e

Commits on Oct 16, 2024

• [scan] support closure (#135602)

  Show description for 0bfa1bf
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  0bfa1bf

  Copy full SHA for 0bfa1bf

• [scan] flatten subgraph output and make subgraph inputs to be a slice (#135601)

  Show description for 819d6b1
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  819d6b1

  Copy full SHA for 819d6b1

• [scan] fix typo in signature and remove wrapper (#135600)

  Show description for 0437a22
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  0437a22

  Copy full SHA for 0437a22

Commits on Sep 6, 2024

• [export] fix placeholder name collision tests by removing map call (#135366)

  Show description for 993b564
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  993b564

  Copy full SHA for 993b564

Commits on Sep 5, 2024

• [cond] fix typo in cond codegen (#134708)

  Show description for 4262755
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  4262755

  Copy full SHA for 4262755

• [hop] preserve metadata in re-tracing hop subgraph by running with interpreter (#135159)

  Show description for 38fead8
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  38fead8

  Copy full SHA for 38fead8

Commits on Aug 30, 2024

• [HOP] fix export x inline_inbuilt_nn_modules (#133731)

  Show description for d261a17
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  d261a17

  Copy full SHA for d261a17

• make make_fx collective test single threaded (#134775)

  Show description for 932c4ca
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  932c4ca

  Copy full SHA for 932c4ca

Commits on Aug 28, 2024

• [hop] require hops to override __call__. (#134352)

  Show description for b07d0a2
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  b07d0a2

  Copy full SHA for b07d0a2

Commits on Aug 24, 2024

• [export] support set_grad_enabled hop in dynamo to enable re-tracing (#134281)

  Show description for 4ff1a4d
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  4ff1a4d

  Copy full SHA for 4ff1a4d

Commits on Aug 23, 2024

• [hop] ban creating hop by directly instantiating HigherOrderOperator. (#133645)

  Show description for a23d86c
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  a23d86c

  Copy full SHA for a23d86c

Commits on Aug 21, 2024

• [hop] ban creating hop by directly instantiating HigherOrderOperator. (#133645)

  Show description for 696107e
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  696107e

  Copy full SHA for 696107e

• [HOP] support generating schema for hop (#133521)

  Show description for 6835f20
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  6835f20

  Copy full SHA for 6835f20

Commits on Aug 19, 2024

• [cond] support unbacked symbool inputs (#133589)

  Show description for 2ec95ff
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  2ec95ff

  Copy full SHA for 2ec95ff

Commits on Aug 9, 2024

• [export] detach constant tensors when they're not registered as buffer or parameter in unlift (#133031)

  Show description for c44cb89
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  c44cb89

  Copy full SHA for c44cb89

Commits on Aug 8, 2024

• Split of "[reland] [export] fix zero arg export in training_ir and constant tensor handling" (#132307)

  Show description for bbf568a
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  bbf568a

  Copy full SHA for bbf568a

Commits on Aug 5, 2024

• [hop] makes compiled hops not share code objects (#132427)

  Show description for a4ed8ee
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  a4ed8ee

  Copy full SHA for a4ed8ee

• fix torch rec test failure (#132437)

  Show description for 618e2c9
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  618e2c9

  Copy full SHA for 618e2c9

Commits on Aug 2, 2024

• [export] change deepcopy to copy in _replace_set_grad_with_hop pass.. (#132181)

  Show description for 19897a1
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  19897a1

  Copy full SHA for 19897a1

Commits on Aug 1, 2024

• [torchbind] don't warning for certain skippable methods. (#132306)

  Show description for 35400f7
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  35400f7

  Copy full SHA for 35400f7

Commits on Jul 30, 2024

• Specialize sym node when used as device kwarg (#131811)

  Show description for 32c57e7
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  32c57e7

  Copy full SHA for 32c57e7

Commits on Jul 26, 2024

• [export] fix set_grad x tensor constant. (#131787)

  Show description for 404a8ae
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  404a8ae

  Copy full SHA for 404a8ae

Commits on Jul 25, 2024

• [export] fix output node's meta (#131706)

  Show description for 2c1851f
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  2c1851f

  Copy full SHA for 2c1851f

• [export][BE] consolidate export and export_for_training (#131496)

  Show description for 75c4176
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  75c4176

  Copy full SHA for 75c4176

• [dynamo] lazily guard and specialize on the symint when used in f-string. (#131529)

  Show description for ffc6bf8
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  ffc6bf8

  Copy full SHA for ffc6bf8

Pagination

Previous

Next