strisunshinewentingwang/pytorch

Pull Requests Commits

Revert D16368004: [pytorch][PR] Fix error message for a wrong fork CUDA

pjh5 committed 5 years ago

06d11f04

Verify flatten works for quantized Tensor (#23121)

jerryzh168 committed 5 years ago

38615206

Fix broken indexing when using None and ellipses indexing together (#22905)

Chillee committed 5 years ago

a24f6c13

Fix load op to return the shape info as before when loading multiple blobs (#23182)

pradeepdfb committed 5 years ago

648f10be

nn_quantized test (#23169)

jerryzh168 committed 5 years ago

1c574458

Turning on fbgemm for nightlies (#22784)

pjh5 committed 5 years ago

e08f8f45

Fix error message for a wrong fork CUDA (#23030)

Guanheng Zhang committed 5 years ago

a6e45a69

Add get_accessed_features function to ModelLayer class (#23036)

kevinwilfong committed 5 years ago

3ca7c0ff

Pin numba to 0.44.0 to fix Windows CI.

ezyang committed 5 years ago

ff23a02a

Remove empty THCThreadLocal{.h/.cpp} (#23157)

vishwakftw committed 5 years ago

b6d06d54

Invert ownership between PyFunction and THPFunction.

ezyang committed 5 years ago

fdfc676e

Support converting Python number to IValue in pybind_utils.h (#22817)

mrshenli committed 5 years ago

ae5b5208

BlackBoxPredictor OSS part 4: Open-source other transforms (#23099)

salexspb committed 5 years ago

2becbd3f

Updating producer_version in exported ONNX models to pytorch 1.2. (#23120)

Spandan Tiwari committed 5 years ago

27031dcc

Fixed deprecated use of yaml.load

Chillee committed 5 years ago

7e31c02a

Refactor named inference rule for reductions

zou3519 committed 5 years ago

76291829

Implement tensor.size(Dimname), tensor.stride(Dimname)

zou3519 committed 5 years ago

b4b51ed5

Bidirectional GRU and LSTM C++ API forward fix (#22850)

pbelevich committed 5 years ago

965b97f5

Revert D16390551: Fix load op to return the shape info as before when loading multiple blobs

ezyang committed 5 years ago

e5797e93

Support get/setstate with no args (#23119)

davidriazati committed 5 years ago

fcdfc35d

Cleanup API and remove 'experimental' warning (#23000)

orionr committed 5 years ago

858d4a6a

Fix type hints for None constants (#23029)

davidriazati committed 5 years ago

fad3031b

Resolve with closed over variables instead of stack frame (#22270)

davidriazati committed 5 years ago

2891784a

Fix load op to return the shape info as before when loading multiple blobs (#23166)

pradeepdfb committed 5 years ago

fd90b967

Added running via throughput benchmark options. (#23077)

kimishpatel committed 5 years ago

82db5dce

Added add op framework overhead benchmark for C2 (#23078)

kimishpatel committed 5 years ago

2ba516d5

Add simple add op based framework overhead benchmark. (#23076)

kimishpatel committed 5 years ago

0621068c

fix qat tests (#23124)

jerryzh168 committed 5 years ago

4223e2f9

Remove cuda free mutex (#23040)

jeffdaily committed 5 years ago

8bc28cc8

izdeby committed 5 years ago

22f7c9e3

Newer Older