strisunshinewentingwang/pytorch

Pull Requests Commits

Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

4ac11bb3

Update base for Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

1c5a84c8

Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

4a93bf39

Update base for Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

09efe42c

Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

67ff3a8b

Update base for Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

cc5e74af

More robust slicing logic for nn.ModuleList (#58361)

tugsbayasgalan committed 3 years ago

d88d321e

[Reducer] Remove replica size == 1 checks (#58603)

rohan-varma committed 3 years ago

b3015584

[DDP] Remove train call to module copies (#58595)

rohan-varma committed 3 years ago

1d67c6d6

[Reducer] move comment to the right place (#58594)

rohan-varma committed 3 years ago

88c76b43

Format reducer.cpp, hpp (#58593)

rohan-varma committed 3 years ago

d83c5a5c

[fx][graph_drawer] Improve graph drawer coloring and tensor_meta handling (#58699)

jfix71 committed 3 years ago

6d97a80d

[codemod][dirsync] Apply clang-format

zertosh committed 3 years ago

5455df2b

Revert D28497967: [quant][fx][graphmode][refactor] Remove qconfig_map from Quantizer

Chillee committed 3 years ago

21a93340

[quant][fx][graphmode][refactor] Remove qconfig_map from Quantizer (#58455)

jerryzh168 committed 3 years ago

1cf8f7a4

[Reducer] Completely remove VariableIndex (#58592)

rohan-varma committed 3 years ago

62adf9e1

[Try] [PyTorch Edge] Trim unused code related to CUDA and HIP Interfaces (#58689)

dhruvbird committed 3 years ago

8e4fc006

Tag PyObject on TensorImpl per torchdeploy interpreter (#57985)

ezyang committed 3 years ago

773cfae9

Change native functions to take `c10::string_view` args instead of `std::string` (#57680)

kurtamohler committed 3 years ago

fe8e5eb2

[Caffe2] [Easy] Fix comment on caffe2_serialize_using_bytes_as_holder to reflect correct types

voznesenskym committed 3 years ago

d1d24304

[Pytorch Edge] NAME -> SCHEMA (#58604)

JacobSzwejbka committed 3 years ago

db67699a

enable torch.cpu.amp.autocast (#57386)

leslie-fang-intel committed 3 years ago

0ede83db

Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

4794c5ec

Update base for Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

6a7fd58b

Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

fd1b8978

Update base for Update on "[c10d] Introduce ProcessGroupWrapper"

rohan-varma committed 3 years ago

5ff82672

[quant][graphmode][fx] Move qat_swap_modules outside of Quantizer (#58454)

jerryzh168 committed 3 years ago

b6dcdeac

[PyTorch] Migrate TI usage in ATen/native/cpu to borrowing (#58303)

swolchok committed 3 years ago

fdc5dfdd

[PyTorch] Add TI::borrowing_nullary_op and use it (#58280)

swolchok committed 3 years ago

7c15d320

Enable the quantization on XPU devices (#54857)

chengjunlu committed 3 years ago

618be18a

Newer Older