strisunshinewentingwang/pytorch

Pull Requests Commits

Merge branch 'add_support_for_optional_tensor' of https://github.com/pytorch/pytorch into add_support_for_optional_tensor

nikithamalgi committed 3 years ago

c07414b9

Merge branch 'add_support_for_optional_tensor' of https://github.com/pytorch/pytorch into add_support_for_optional_tensor

nikithamalgi committed 3 years ago

77a7cfa5

Merge branch 'add_support_for_optional_tensor' of https://github.com/pytorch/pytorch into add_support_for_optional_tensor

nikithamalgi committed 3 years ago

c102a94e

Resolve merge conflicts

nikithamalgi committed 3 years ago

37337e00

Resolve merge conflicts

nikithamalgi committed 3 years ago

0cbc08b7

Merge branch 'master' of https://github.com/pytorch/pytorch into add_support_for_optional_tensor

nikithamalgi committed 3 years ago

f6bceac5

Annotate NoneType as Optional[torch.Tensor]

nikithamalgi committed 3 years ago

902edded

Annotate NoneType as Optional[torch.Tensor]

nikithamalgi committed 3 years ago

06cf7ed6

Remove unused sample input argument from functions to resolve issue #55737 (#60486)

NivekT committed 3 years ago

a2525b03

Add device runtime API for the plug-in to register platform python module into torch (#59857)

chengjunlu committed 3 years ago

265f0e53

Fix test failures with some glibc libraries (#60450)

Flamefire committed 3 years ago

c97d4d5a

Clean Up ZeRO (#60285)

awgu committed 3 years ago

f0e4e4be

Ensure proper syncs for out-of-place grad creation (torch.autograd.grad) when backward ops run on side streams (#60127)

mcarilli committed 3 years ago

56481f97

Revert D29190420: [nnc][tests] Tests and benchmarks for computeSum

anjali411 committed 3 years ago

b14f19b6

To add edge_order=2 and documentation for gradient operator (#58165)

iramazanli committed 3 years ago

90cd57ee

[NormalizeArgs] Retain node.meta (#60449)

jfix71 committed 3 years ago

7ed07e2a

Ensure num_threads is initialized before calling omp_get_max_threads (#60185)

peterbell10 committed 3 years ago

66452e0a

OpenMP: Refactor parallel_reduce to share code with parallel_for (#60184)

peterbell10 committed 3 years ago

19553438

Ensure thread id is valid in nested parallel regions (#60183)

peterbell10 committed 3 years ago

c75714e5

Migrate crossKernel from THC to ATen (CUDA) (#60039)

peterbell10 committed 3 years ago

3f3fd570

[BE] Fix Convolution.cpp build warnings (#60463)

malfet committed 3 years ago

f590ccea

Increase tolerance for test_grad_scaling_clipping (#60458)

Flamefire committed 3 years ago

3846cef2

`topk` on CUDA supports `bfloat16` (#59977)

eqy committed 3 years ago

40de03fc

[nnc][tests] Tests and benchmarks for computeSum (#60160)

bertmaher committed 3 years ago

21479ad2

[nnc] Speed up batchnorm benchmark

bertmaher committed 3 years ago

fbeb8b49

[pytorch][nnc] external function call to xnnpack ops (#59525)

ljk53 committed 3 years ago

b0c9762e

Add error message for sequence length to be equal to 0 case for RNNs (#60269)

iramazanli committed 3 years ago

79dc500a

Add custom code filter for TS (#60309)

nikithamalgi committed 3 years ago

dc9aa7b9

[quant] Input-Weight Equaliaztion - convert modifications (#59963)

angelayi committed 3 years ago

3de79b77

Enable rcb lookup for typing (#60413)

nikithamalgi committed 3 years ago

7589d9c5

Older