strisunshinewentingwang/pytorch

Pull Requests Commits

Valentine233 committed 2 years ago

b74d49b0

add inductor debug tool

Valentine233 committed 2 years ago

bfab07a5

add inductor debug tool

Valentine233 committed 2 years ago

55419c37

inductor: support rsqrt for dynamic shape (#103579)

XiaobingSuper committed 2 years ago

da21273a

Handle long Docker image name when building Docker image (#103562)

huydhn committed 2 years ago

5efdcd58

enable more ASAN tests (#101483)

cyyever committed 2 years ago

1e108d9c

Inductor cpp wrapper: support Constant in input (#103496)

chunyuan-w committed 2 years ago

17217d36

[PT2][Quant] Update op names for decomposed quantized lib (#103251)

kimishpatel committed 2 years ago

90ee6a73

cm3leon_generate is at edge of timeout, so bump it up (#103607)

ezyang committed 2 years ago

5211fad7

chore: Update ModelReportObserver variables to buffers (#97971)

pkluska committed 2 years ago

b4056ba7

Register more foreach op lowerings (#102654)

mlazos committed 2 years ago

00546333

tf32 context fixes for various tests (#103137)

Fuzzkatt committed 2 years ago

6d570ccd

Fix inductor-perf-compare (#103538)

peterbell10 committed 2 years ago

2e653548

Revert "[inductor] Make clone_graph copy node name as well (#103409)"

pytorchmergebot committed 2 years ago

3d6fd07c

[benchmark] hf_T5_base - torchbench original batchsize too large (#103442)

anijain2305 committed 2 years ago

d6da649a

[benchmark][compile] Limit number of bounding boxes to 5 (#103413)

anijain2305 committed 2 years ago

16c2090b

Revert "Support params/buffers inside cond and map (#102310)"

pytorchmergebot committed 2 years ago

2087d328

Delete ifdyn and ifunspec combinators (#103596)

ezyang committed 2 years ago

ddf4cd69

Add `generator` argument in `torch.randn` signature (#102075)

liang-hou committed 2 years ago

e82616d9

Link torch.cat in docstring of torch.stack and vice versa (#103421)

moi90 committed 2 years ago

a0885dff

Support params/buffers inside cond and map (#102310)

tugsbayasgalan committed 2 years ago

766f236b

add instruction to compile with new C++ ABI (#95177)

jingxu10 committed 2 years ago

600f7dc2

Improve DDPOptimizer Logging (#103489)

wconstab committed 2 years ago

55cf5c00

Silence `has_cuda` deprecation in optim (#103610)

Andrew Gu committed 2 years ago

9152d0e5

[Pytorch] aten::stack (#103344)

lucylq committed 2 years ago

d0ff640e

Fix composable `checkpoint(use_reentrant=True)` with multi args (#103590)

Andrew Gu committed 2 years ago

2eea3cb1

[inductor] Fix an expression printer issue during generate_return (#103557)

desertfire committed 2 years ago

c2952e8b

Update optimizer tests to compile with fullgraph (#103559)

mlazos committed 2 years ago

dc3fa9e5

[FSDP][4/n]Update use_dtensor option for _optim_utils.py (#103599)

wz337 committed 2 years ago

7dd0f525

[activation checkpoint][dynamo] Wrap AC into Tag based higher order op (#102935)

anijain2305 committed 2 years ago

bd0ed940

Older