strisunshinewentingwang/pytorch

Pull Requests Commits

Add `is_causal` API for `TransformerDecoder` (#97166)

janEbert committed 1 year ago

95c166cd

Support per channel tensor with unpacking in QNNPACK (#96268)

kwanghoon-meta committed 1 year ago

92605ee7

[DataPipe] Fix missing imports in DataPipe interface file (#97458)

NivekT committed 1 year ago

c5135ff2

Warn once on dynamo module w/ hooks (#97535)

wconstab committed 1 year ago

827b2aee

[Kineto] Improve Config Options for Input Shapes, Memory, Stack, Flops, and Modules - Part 1 (#97380)

aaronenyeshi committed 1 year ago

197434df

Use L1 loss for Smooth L1 loss with beta=0 (#97022)

CedricPicron committed 1 year ago

cf0ba1b9

[pytorch@arvr/windows] Fix pytorch build/import on Windows @ ovrsource (#97193)

esauser2 committed 1 year ago

17567e5b

[ROCm] Update clock intrinsic handling for AMD gfx11 family (#97005)

pruthvistony committed 1 year ago

baf71a8a

Revert "Upgrade NVTX to NVTX3 (#90689)"

pytorchmergebot committed 1 year ago

5170995b

Code update for vectorized interpolate cpu uint8 (#96847)

vfdev-5 committed 1 year ago

a96ccaa3

backport std::ssize to c10 (#97442)

mikey dagitses committed 1 year ago

4ff71c91

`GradScaler` recomputes `optimizer_state["found_inf_per_device"]` before `optimizer.step` (#97415)

crcrpar committed 1 year ago

b5edf183

Complex support for expm1 (#96644)

yhl48 committed 1 year ago

6fcd6715

[ROCm] Update magma commit for ROCm (#97491)

pruthvistony committed 1 year ago

1b8b82f8

[CI] Experiment with a newer CUDA driver (#96904)

desertfire committed 1 year ago

c55d1a60

Fix typos under torch/utils directory (#97516)

kiszk committed 1 year ago

622a11d5

[Dynamo] Fix TIMM benchmark compute_loss (#97423)

yanboliang committed 1 year ago

d305d4a5

remove unused CAFFE2_VERSION macros (#97337)

mikey dagitses committed 1 year ago

5f5d6755

Log FSDP mixed precision (#97367)

rohan-varma committed 1 year ago

605a77fd

[ONNX] Support converting fx graph with symbolic shape to ONNX (#96350)

titaiwangms committed 1 year ago

51ce0223

add /std:c++17 to windows compilations when not using Ninja (#97445)

mikey dagitses committed 1 year ago

bcff4773

[inductor] xfail tests by default (#97331)

nkaretnikov committed 1 year ago

6e46f472

Disable inductor developer warnings in official releases (#97451)

msaroufim committed 1 year ago

36d64760

Use a single stream for cuda graph pool (#97419)

eellison committed 1 year ago

73fadd52

Bring back tensor_has_compatible_shallow_copy_type (#97455)

fduwjj committed 1 year ago

b11ce4bb

Revert "Rewrite NCCL watchdog to more reliably throw timeout (#97066)"

pytorchmergebot committed 1 year ago

f25cdf8a

[Sparse] Add reference implementation for addmv (#97353)

malfet committed 1 year ago

ad5d81ad

Add missing aot_autograd_arg_pos_to_source (#97487)

voznesenskym committed 1 year ago

31e858e8

Add GPU frequency lock option to inductor workflows running on A100 (#97465)

weiwangmeta committed 1 year ago

9320cae1

Rename PyOperator to HigherOrderOperator (#97493)

ezyang committed 1 year ago

fa4c77e3

Newer Older