strisunshinewentingwang/pytorch

Pull Requests Commits

Update jit_language_reference_v2.rst

brianjo committed 3 years ago

Verified 13fe8dff

Revert D26676150: Simplify index expressions constructed in loop flattening - #51173

ezyang committed 3 years ago

07d315fc

Adding parallel support for the LLVM backend. (#53243)

zheng-xq committed 3 years ago

95d23185

[JIT] Update set_stream API to change the device (#53741)

nikithamalgifb committed 3 years ago

351f6f5e

[JIT] Update Namespace from cuda to _cuda (#53378)

nikithamalgifb committed 3 years ago

cfaa0bf2

Perform appropriate CUDA stream synchronization in distributed autograd. (#53769)

pritamdamania committed 3 years ago

0b84f45f

[GraphModule] Back out changes to module root version of __init__ (#53791)

jfix71 committed 3 years ago

1053c966

Adding learning rate schedulers to C++ API (#52268)

jamesbut committed 3 years ago

37ab7118

Move prim::layout for lite jit (#53781)

jiatongzhou committed 3 years ago

ebfa9276

[nnc] Test ability to vectorize reads from an intermediate tensor (#53752)

bertmaher committed 3 years ago

3bd250fd

[NNC] LoopNest cleanup (#53688)

navahgar committed 3 years ago

a5e19126

[docs] Add starter content for new TorchScript language reference (#52494)

Meghan Lele committed 3 years ago

d49c5c74

[deploy] namespace and rename (#53670)

zdevito committed 3 years ago

ced91bb7

[PyTorch] Speed up Tensor::data_ptr (#53723)

swolchok committed 3 years ago

14acf92b

Simplify index expressions constructed in loop flattening - #51173 (#52882)

huiguoo committed 3 years ago

1f01899e

[nnc] Add an initialization expression to Reduce() (#53751)

bertmaher committed 3 years ago

997f05cd

skip dispatch in resize_ (#53575)

bdhirsh committed 3 years ago

49a5f994

Avoid of creating a copy of statusString every inference time (#53756)

hl475 committed 3 years ago

21f9a6da

[quant][fx][graphmode][fix] Only insert observers for fixed qparam ops (#53330)

jerryzh168 committed 3 years ago

0584fd93

[quantization] Add some support for 3d operations (#50003)

dmenig committed 3 years ago

f9185973

TensorIterator: Avoid nesting two levels of function_ref in for_each (#53613)

peterbell10 committed 3 years ago

895735c6

Add a section to introduce GradBucket class in ddp_comm_hooks.rst (#53253)

Yi Wang committed 3 years ago

fe0810e2

Add a description of GradBucket Python class (#53596)

Yi Wang committed 3 years ago

c988b78b

[package] split tests (#53749)

suo committed 3 years ago

741d0f41

Fix error message in setStorage (#53198)

zdevito committed 3 years ago

4351d096

Remove trailing whitespace introduced in #52175 (#53762)

samestep committed 3 years ago

fee26359

[caffe2] update load_save_test.py to also verify the chunking behavior (#53401)

simpkins committed 3 years ago

023948e6

[caffe2] use AddNAlreadyReserved() when serializing blobs (#53400)

simpkins committed 3 years ago

99d7c8ff

[TestZeroRedundancyOptimizer] Add multi gpu checker (#53564)

jaglinux committed 3 years ago

2cf90982

[quant][graphmode][fix] Handle the case when observed node has no users (#53210)

jerryzh168 committed 3 years ago

d9fa957e

Older