strisunshinewentingwang/pytorch

Pull Requests Commits

add missing NEON {vld1,vst1}_*_x2 intrinsics (#43683)

Sebastian Pop committed 4 years ago

c2591464

Back out "Selective meta programming preparation for prim ops"

Jae Lee committed 4 years ago

137a4fcc

Rename is_complex_t -> is_complex (#39906)

zasdfgbnm committed 4 years ago

263412e5

[TensorExpr] Remove unused functions in kernel.cpp (#43966)

asuhan committed 4 years ago

9db90fe1

[quant][graphmode][fx] Support dynamic quantization without calibration (#43952)

jerryzh168 committed 4 years ago

8fd9fe93

broadcast_object API for c10d (#43887)

rohan-varma committed 4 years ago

fbea2ee9

Pass CC env variable as ccbin argument to nvcc (#43931)

malfet committed 4 years ago

4134b7ab

[quant][graphmode][fx] Support dynamic quantization without calibration (#43892)

jerryzh168 committed 4 years ago

0ffe3d84

[quant][graphmode][fx][refactor] Move patterns to separate files (#43891)

jerryzh168 committed 4 years ago

d15b9d98

[FX] Better error when unpacking Proxy (#43740)

James Reed committed 4 years ago

8d53df30

[OSS] Update README.md -- Explain more complex arguments and functionalities

yujunzhao@devvm537.prn0.facebook.com committed 4 years ago

ec7f1494

Delete `raise_from` from `torch._six` (#43981)

malfet committed 4 years ago

e49dd9fa

Enable TF32 support for cuDNN (#40737)

zasdfgbnm committed 4 years ago

5e97f251

Update `README.md` in oss (#43893)

yujunzhao@devvm1621.atn0.facebook.com committed 4 years ago

93fbbaab

Check SparseAdam params are dense on init (#41966) (#43668)

Randall Hunt committed 4 years ago

24eea364

Selective meta programming preparation for prim ops (#43540)

iseeyuan committed 4 years ago

bacee6aa

[FX] Pickle serialization of GraphModule via forward source (#43674)

James Reed committed 4 years ago

a1a23669

[FX] Support tensor-valued constants (#43666)

James Reed committed 4 years ago

73f7d63b

[TVM] Support slice op (#43969)

hlu1 committed 4 years ago

06c277f3

Reset `DataLoader` workers instead of creating new ones (#35795)

Emilio Castillo committed 4 years ago

5472426b

rename input argunment `interested-folder` to `interest-only` -- be consistent with other arguments (#43889)

yujunzhao@devvm1621.atn0.facebook.com committed 4 years ago

db6bd9d6

Back out "Revert D19987020: [pytorch][PR] Add the sls tensor train op" (#43938)

lly-zero-one committed 4 years ago

bc64efae

Revert D23216393: Support work.result() to get result tensors for allreduce for Gloo, NCCL backends

malfet committed 4 years ago

7035cd0f

Add typing annotations for torch.nn.quantized.dynamic.modules.rnn (#43186)

guilhermeleobas committed 4 years ago

63a0bb0a

Introduce BUILD_CAFFE2 flag (#43673)

Rong Rong committed 4 years ago

8ca3913f

[pytorch][bot] update mobile op deps (#43937)

ljk53 committed 4 years ago

76ca3656

Error printing extension support for multiline errors (#43807)

Lilyjjo committed 4 years ago

e3cb582e

Move Autograd to an alias dispatch key (#43070)

Ailing Zhang committed 4 years ago

22423203

MaxPool1d without indices optimization (#43745)

heitorschueroff committed 4 years ago

13a48ac1

updated documentation to streamline setup (#42850)

osandoval-fb committed 4 years ago

a044c039

Newer Older