strisunshinewentingwang/pytorch

Pull Requests Commits

Fix dead link and syntax in ONNX landing page

BowenBao committed 5 years ago

bbf84c1a

Attempt to fix windows build

James Reed committed 5 years ago

0c222555

Updating submodules

svcscm committed 5 years ago

d49f1349

Some alias analysis fixes (#25425)

suo committed 5 years ago

194acd02

neginraoof committed 5 years ago

d2919353

Momentum setting in SyncBatchNorm forward (inference) pass. (#24995)

Spandan Tiwari committed 5 years ago

8986b9e3

Quantized vec256 + vectorized quantized::add

James Reed committed 5 years ago

17831648

relax roi_width/roi_height check to non-negative

wat3rBro committed 5 years ago

8cd45b4c

Attempt to enable CrossMapLRN2d, as it no longer uses Module._backend.

gchanan committed 5 years ago

93b653bb

Change exception to warning (#25408)

Haixin Liu committed 5 years ago

c59540b7

add speed benchmark binary for torch jit (#25230)

ljk53 committed 5 years ago

86b1d5f2

fix binaries build for BUILD_CAFFE2_MOBILE=OFF (#25229)

ljk53 committed 5 years ago

e370486d

Assign each RpcAgent a unique ID, and use ID for sending RPC messages. (#24195)

mrshenli committed 5 years ago

1294e55c

Remove unnecessary checks in InsertQuantDeQuantImpl (#25370)

jerryzh168 committed 5 years ago

629a2b36

Skip inserting observers for Tensors inside fused op (#25281)

jerryzh168 committed 5 years ago

f495a3ab

Per Channel Quantization Support for Quantized Linear Operator (#25276)

jianyuh committed 5 years ago

88a27ebb

Skip test_compare_tensor_scalar due to overflow error (#25432)

supriyar committed 5 years ago

3805be62

Update QNNPACK submodule to 7d2a4e9 (#25400)

supriyar committed 5 years ago

a9bb68d4

Delete toType(const DeprecatedTypeProperties&, ...) (#25332)

ezyang committed 5 years ago

7b4edded

Add Int8Transpose operator (#16382)

dreiss committed 5 years ago

d704097d

making quant utilities inplace

z-a-f committed 5 years ago

e44c09ec

Remove Module._backend as it's not used anymore.

gchanan committed 5 years ago

23fde77d

Revert D17078081: Invariant typevar matching on callsite checks

wanchaol committed 5 years ago

f077847a

Revert D17003555: Multiple fixes to test_c10d.py.

Mingbo Wan committed 5 years ago

247cac26

Fix allreduce_coalesced tests in c10d (#25419)

mrshenli committed 5 years ago

04764d57

Add guards for using named tensor with serialization and multiprocessing (#25345)

zou3519 committed 5 years ago

2513ca66

Add guard for named tensors in the JIT (#25344)

zou3519 committed 5 years ago

0bb69f60

Add support for non-affine batch norm with float stats and half inputs (#22750)

ptrblck committed 5 years ago

8640aef5

Fix `item()` call in docs

davidriazati committed 5 years ago

fe922a2e

Fixed masking warnings in tests (#25317)

izdeby committed 5 years ago

1ea1d7f0

Newer Older