Commits

Branch selector

 main

User selector

mrshenli

Datepicker

All time

Commit History

Commits on Aug 15, 2023

• [Reland] Add OnCompletion Hook to ProcessGroup (pytorch#106988) (pytorch#107233)

  Show description for 45128ab
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  45128ab

• Add OnCompletion Hook to ProcessGroup (pytorch#106988)

  Show description for ba1da47
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  ba1da47

Commits on Aug 14, 2023

• Add distributed/c10d *.hpp files to lintrunner (pytorch#107160)

  Show description for 064d813
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  064d813

• [Easy] Make Work::retrieveOpType a const function (pytorch#107141)

  Show description for facadc6
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  facadc6

• Apply clang-format to distributed/c10d folder (pytorch#107140)

  Show description for dd63191
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  dd63191

Commits on May 15, 2023

• [SPMD] Allow Override.replacement to have a global view (pytorch#101427)

  Show description for af841f3
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  af841f3

Commits on May 3, 2023

• [SPMD] add FQN argument to Override.replacement (pytorch#100473)

  Show description for 2ebb48f
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  2ebb48f

• Avoid using einsum for torch.cat DTensor propogation (pytorch#100251)

  Show description for 9439cb0
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  9439cb0

Commits on Apr 19, 2023

• [SPMD][Easy] switch to tree_map_only to simplify code (pytorch#99547)

  Show description for ca89e79
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  ca89e79

• [SPMD] Add sym_stride to DSymInt (pytorch#99504)

  Show description for e605b5d
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  e605b5d

• [SPMD] Support DSymInt for slice_backward in SPMD expansion (pytorch#99501)

  Show description for 2cb8a8d
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  2cb8a8d

• [SPMD] Support SymInt with non-op call_function nodes (pytorch#99420)

  Show description for 2922961
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  2922961

• [SPMD] Add aten.stack and aten.select to DTensor prop (pytorch#99417)

  Show description for 7c0c663
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  7c0c663

• Avoid import * from experimental_ops (pytorch#99363)

  Show description for 301be37
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  301be37

Commits on Apr 17, 2023

• [SPMD][Easy] Making typing consistent by replacing object with Any (pytorch#99332)

  Show description for 62a6d81
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  62a6d81

• [SPMD][EASY] Remove unnecessary torch.ops prefix (pytorch#99331)

  Show description for 19c2804
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  19c2804

Commits on Apr 16, 2023

• [SPMD][BE] Generalize factory ops support in SPMD expansion (pytorch#99233)

  Show description for c69d548
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  c69d548

• [SPMD][BE] Remove deprecated aten.sym_numel branch (pytorch#99232)

  Show description for 6bb2082
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  6bb2082

• [SPMD][BE] Consolidate DSymInt Branches (pytorch#99231)

  Show description for 39be994
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  39be994

• [SPMD] Refactor DSize to DSymInt to enable sym_numel (pytorch#99206)

  Show description for 544cd8e
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  544cd8e

• [SPMD] Enable aten.full.default with SymInt on sharded dims (pytorch#99190)

  Show description for bafb984
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  bafb984

Commits on Apr 14, 2023

• Respect sharded dimensions when aten expaned/view consumes SymInt values (pytorch#99058)

  Show description for 40aaacd
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  40aaacd

• [Easy] Clean up args remap for DTensor expansion (pytorch#99040)

  Show description for 02d1cf5
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  02d1cf5

Commits on Apr 13, 2023

• Support FQN as SPMD module override key (pytorch#98966)

  Show description for 75f55ca
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  75f55ca

Commits on Apr 10, 2023

• Avoid passing buffers to optimizers during spmd rematerialization (pytorch#98714)

  Show description for 3fcc5ff
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  3fcc5ff

Commits on Apr 9, 2023

• Support LayerNorm without weight or bias parameters (pytorch#98687)

  Show description for 54b1684
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  54b1684

• Enable replicated embedding in SPMD for NLP models (pytorch#98686)

  Show description for 1be3549
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  1be3549

Commits on Apr 8, 2023

• Add NLLLoss to DTensor prop rule (pytorch#98512)

  Show description for d255c8e
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  d255c8e

Commits on Apr 7, 2023

• Fix wrong SPMD test target in test_log_softmax (pytorch#98610)

  Show description for b2b783e
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  b2b783e

Commits on Apr 6, 2023

• Remove unnecessary schema_map from spmd API (pytorch#98444)

  Show description for 0c5389b
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  0c5389b

• [Easy] remove unnecessary get_rank() in tests (pytorch#98445)

  Show description for 0199140
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  0199140

• [SPMD] Dedup collectives from DTensor expansion (pytorch#98216)

  Show description for 1cd1d9c
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  1cd1d9c

• Enable LogSoftmax for SPMD tracing (pytorch#98380)

  Show description for 11b0a84
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  11b0a84

Commits on Apr 5, 2023

• [Easy] Fix minor errors in DTensor examples (pytorch#98430)

  Show description for d0eafed
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  d0eafed

Commits on Apr 2, 2023

• [Easy] Fix lint error on DTensor math_ops.py (pytorch#98170)

  Show description for 96403cf
  

  mrshenli

  authored and

  pytorchmergebot

  committed
  96403cf

Pagination

PreviousNext