strisunshinewentingwang/pytorch

Pull Requests Commits

Make JIT Aliasing Test Less Brittle (#65493)

Jiewen Tan committed 2 years ago

6502fb89

[QPL] move metadata logging to markerEnd for model run QPL (#65451)

linbinyu committed 2 years ago

4f5ea598

[caffe2] fix LLVM-12 nullptr-with-nonzero-offset UBSAN error (#65506)

Igor Sugak committed 2 years ago

2481c064

Migrate THCCachingHostAllocator to ATen (#65746)

peterbell10 committed 2 years ago

f6dfac69

[ONNX] Enable export of __xor_ (#64042) (#64581)

BowenBao committed 2 years ago

d3979034

[ONNX] Fix inplace fill_ dtype export mismatch (#64233) (#64580)

BowenBao committed 2 years ago

e598ba2e

[ONNX] Update doc and error message for indexing export (#64290) (#64579)

BowenBao committed 2 years ago

89cbe622

[ONNX] Fix remainder export (#64230) (#64578)

BowenBao committed 2 years ago

d4ff344f

Add onnx test for batched_nms (#53175) (#64381)

BowenBao committed 2 years ago

0f0ef4fe

[ONNX] Fix gather squeeze axis in constant folding (#63588) (#64379)

BowenBao committed 2 years ago

7e15f2dd

[ONNX] Fix cuda test case (#63597) (#64378)

BowenBao committed 2 years ago

41bdfe39

[ONNX] Fix input sequence for pad op (#60554) (#64377)

BowenBao committed 2 years ago

2d61009f

Fix empty size constant creation (#63607) (#64376)

BowenBao committed 2 years ago

f17ee368

[ONNX] Update instance_norm implementation and support training (#60538) (#64375)

BowenBao committed 2 years ago

84190daf

[fx2trt][quant] Add lowering support for per channel quantization in fx2trt (#64787)

jerryzh168 committed 2 years ago

3d6d4f43

Delete rouge cu102 windows builds (#65961)

malfet committed 2 years ago

207fefc9

Clarified difference in behavior of `empty_strided` and `as_strided` (#64568)

ysiraichi committed 2 years ago

b3da2afe

Revert D31137652: [pytorch][PR] Skip failing tests when LAPACK and MAGMA are not available

suo committed 2 years ago

22f36353

Implement n-dimensional hermitian FFTs (#63890)

peterbell10 committed 2 years ago

6285348f

Add __module__ to torch.dtype.__dict__ (#65182)

nelhage committed 2 years ago

70f9f58a

[PyTorch][Edge] Fix inefficiency in objLoaderMobile (#65710)

swolchok committed 2 years ago

38c77539

[MicroBench] Added a micro benchmark for prefix sum (#65790)

navahgar committed 2 years ago

8f398325

[ci] fix softmax bc check (#65952)

suo committed 2 years ago

24f59fa2

Refactor `OperatorSupport` related code and fix TRT not supporting int64 dtype (#65848)

kflu committed 2 years ago

d4d3bb91

Revert D31238123: [pytorch][PR] Avoid saving self for`softmax` and `log_softmax`

suo committed 2 years ago

9ae63bd8

Add cuSPARSE descriptors and update CSR addmm (#60838)

IvanYashchuk committed 2 years ago

541eb1db

Update git version for CentOS base dockers (#65703)

jithunnair-amd committed 2 years ago

be00f020

[ci] try installing libgnutls to fix cert error (#65934)

suo committed 2 years ago

8297a16c

Move ASAN to GHA (#65846)

malfet committed 2 years ago

6a30d835

.github: Bump linux and windows gpu max available (#65923)

seemethere committed 2 years ago

cdbfb2b6

Newer Older