strisunshinewentingwang/pytorch

Pull Requests Commits

[torch][elastic] Make final agent barrier to shutdown properly

aivanou committed 2 years ago

e54c1f6c

[2] remove caffe2 math.h from maskrcnn ops

linbinyu committed 2 years ago

bc72add5

[lint] Use a problem matcher for GitHub annotations

suo committed 2 years ago

85235c6f

[lint] add test ownership lint to lintrunner

suo committed 2 years ago

3e0e1375

set fetch-depth: 1

clee2000 committed 2 years ago

991c89b2

Revert "Add support to `Tensor[]?` for structured kernel codegen."

malfet committed 2 years ago

3c238c6c

[NVFuser] fix incorrect vector size in chunk move

davidberard98 committed 2 years ago

15892481

[lint] improve retries in stale job

suo committed 2 years ago

9555f3b3

[jiterator] logical_and: complex

khushi-411 committed 2 years ago

fa2c7e2e

Move //xplat/caffe2:caffe2_serialize to shared build structure

kit1980 committed 2 years ago

a5b4839f

Remove spurious warning when using disabled torch function

albanD committed 2 years ago

3467f3fa

Add support to `Tensor[]?` for structured kernel codegen.

ysiraichi committed 2 years ago

ac2de5a0

Utilities for handling Sparse Compressed tensors

pearu committed 2 years ago

0ab03af5

Fix data race on owns_pyobj_ accesses with non-GIL protected threads

ezyang committed 2 years ago

337e3932

Set correct module for APIs in torch module

anjali411 committed 2 years ago

533f0cb2

Preserve Python dispatch keys upon copy_tensor_metadata_except_version_counter

ezyang committed 2 years ago

27728708

Improves the OpInfo out= tests

Mike Ruberry committed 2 years ago

b0976999

[lint] use python to run flake8 and mypy in linter

suo committed 2 years ago

123297a8

[lint] add actionlint to lintrunner

suo committed 2 years ago

356f1478

[lint] add shellcheck to lintrunner

suo committed 2 years ago

cbbb96c2

Add layout member to SparseCsrTensorImpl

pearu committed 2 years ago

63b0e1fa

Introduce sparse compressed layouts: SparseCsr, SparseBsr, SparseBsc

pearu committed 2 years ago

1cd46b30

Introduce virtual layout_impl method to TensorImpl

pearu committed 2 years ago

4c872835

[quant][core][gpu][bug fix] Changed at::contiguous call to at::to for quantized cudnn operators

dzdang committed 2 years ago

ef640355

[quant][core][bug fix] Corrected at::to(memory_format=...) support for quantized tensors

dzdang committed 2 years ago

515d61f2

[quant][core][bug fix][gpu] Added kReluFused to quantized cudnn conv operator's caching

dzdang committed 2 years ago

cf735d89

[lint] use new merge-base-with feature in lintrunner ci

suo committed 2 years ago

03cbdaee

Update Index.rst to add TorchRec to domain list.

brianjo committed 2 years ago

990d155c

Support noncontiguous inputs for torch.distributed.nn.functional.all_gather/reducescatter/gather

SherlockNoMad committed 2 years ago

752ab799

[lint] add back some omitted clangtidy folders

suo committed 2 years ago

d2e7e4e2

Newer Older