strisunshinewentingwang/pytorch

Pull Requests Commits

[iOS GPU] Add debug information to track memory allocation exception (#59112)

xta0 committed 3 years ago

ab372ba5

CUDA support in the CSR layout: sparse_to_dense/add_sparse_csr (#59011)

aocsa committed 3 years ago

41054f2a

Use some c10::ThreadLocal to avoid crashes on old Android toolchains (#59017)

dreiss committed 3 years ago

9c83e416

Include Macros.h in ThreadLocal

dreiss committed 3 years ago

4b3d17c0

OpInfo: `fmod` and `remainder` (#57941)

krshrimali committed 3 years ago

0c1420aa

Revert D28700259: [pytorch][PR] Migrate nonzero from TH to ATen (CPU)

Natalia Gimelshein committed 3 years ago

657b75d1

Move remaining \*Sort\* in `THC` to `ATen` (#58953)

eqy committed 3 years ago

4e543d01

Add peephole for len(x.size()) (#59051)

eellison committed 3 years ago

f3aa61b9

Add more shape functions for mobilenet (#58932)

eellison committed 3 years ago

b9dc5186

Switch symbolic shape registry to operator map' (#58890)

eellison committed 3 years ago

0ebc6653

[JIT] Disable Complete Shape Inlining For Testing Purposes (#56966)

eellison committed 3 years ago

d8cbba3e

Add unary/binary ops necessary for mobilenet (#56828)

eellison committed 3 years ago

f66fbb1e

Use dataclasses to simplify ShardingSpec (#58893)

pritamdamania committed 3 years ago

40f851c5

[quant][refactor tests] Move qtensor serialization tests from test_deprecated_jit (#59089)

supriyar committed 3 years ago

89d78851

[quant][refactor tests] Clean up test_quantization.py (#59088)

supriyar committed 3 years ago

886a2ddc

TensorIteratorReduce: Avoid tensor operations in parallel_for (#58655)

peterbell10 committed 3 years ago

f993ceff

Back out "[pytorch][PR] ENH Adds dtype to nn.functional.one_hot" (#59080)

jbschlosser committed 3 years ago

ef32a29c

Retiring _module_copies field in DDP reducer. (#59094)

semrefb committed 3 years ago

3d2b5555

Added minor fixes to Az DevOps Build Logic (#59016)

mstfbl committed 3 years ago

c6c563fc

don't copy indices to the self device in dispatch_index (#59059)

Natalia Gimelshein committed 3 years ago

61f946bb

Revert D28615349: [pytorch][PR] Add a ci/no-build label

samestep committed 3 years ago

16ae6cad

Add a ci/no-build label (#58778)

samestep committed 3 years ago

bae06a02

[docs] document dim argument to tensor.size() (#58777)

garymm committed 3 years ago

3e2db56d

Add script to cancel workflows (#59019)

driazati committed 3 years ago

18302bcd

[PyTorch] Save a refcount bump in meta functions for addmm and mm (#59063)

swolchok committed 3 years ago

920619dc

[ONNX] Enable support for roll() op. (#58389) (#58697)

BowenBao committed 3 years ago

2c17b6a0

[ONNX] handle aten::_set_item on Dict in convertInplaceOpsAndTrackAlias (#58317) (#58696)

BowenBao committed 3 years ago

1aabb8f9

[ONNX] use consistent quoting for string literals (#57757) (#58695)

BowenBao committed 3 years ago

0a6828a3

[ONNX] Improve lower tuples and handle control flow (#57650) (#58694)

BowenBao committed 3 years ago

b27fc0ff

[ONNX] Update special post process for SequenceInsert after SequenceEmpty (#56965) (#58693)

BowenBao committed 3 years ago

57c9355a

Newer Older