strisunshinewentingwang/pytorch

Pull Requests Commits

making import_module private and deprecating public method (#67990)

jdsgomes committed 2 years ago

577a4d34

Removes unnecessary `no_pretrained_model` from test_quantize_fx.py (#67836)

datumbox committed 2 years ago

0a9cd6d4

Fix deadlock for multi-output forward AD (#67995)

albanD committed 2 years ago

f9422e1c

Adds a `maximize` flag to SGD. (#67847)

oliver committed 2 years ago

f8297d40

Disable TF32 in `pinv_jvp` and `pinv_backward` (#67948)

crcrpar committed 2 years ago

c5e5264b

Revert D32007691: [pytorch][PR] Op info for activation functions 2 (softsign, tanh, tanhshrink, threshold, celu, sigmoid, mish, hardsigmoid)

ngimel committed 2 years ago

417dc7f8

Enforce that test cases extend from correct TestCase (#67819)

janeyx99 committed 2 years ago

36d9a74b

Fix typo grid_sampler_3d_cuda (#67752)

Leo7654 committed 2 years ago

25cd8187

[WIP][resubmit] Don't #define NUM_THREADS (#68008)

zasdfgbnm committed 2 years ago

4b1d0444

Fixes CUDA vs CPU consistency for index_put_ when accumulating (part 2) (#67189)

vfdev-5 committed 2 years ago

a2ab0651

[distributed] Render `torch.distributed.optim` members (#67885)

James Reed committed 2 years ago

3f048c63

[fx2trt] fix import in oss tests (#68016)

842974287 committed 2 years ago

fd198a2f

[fx2trt]organize converter utils (#68015)

842974287 committed 2 years ago

0d8a8a2e

[Doc] [ONNX]Fix a broken url for ONNXRuntime custom op (#67944)

jcwchen committed 2 years ago

5b036d5f

Upgrade and fix boto3 version to 1.19.12 (#68025)

janeyx99 committed 2 years ago

82398e38

use better secrets for upload labels workflow (#68013)

janeyx99 committed 2 years ago

9094947b

[ROCm] Bump magma source to pickup memory leak fix (#67225)

dllehr-amd committed 2 years ago

db9b4f1a

[hackathon][DataPipe] adding .pyi file generation for torch.utils.data.datapipes (#67374)

NivekT committed 2 years ago

0b09d62c

[JIT] additional support for CallMethod with autocasting (#67925)

davidberard98 committed 2 years ago

2e523ed2

[ONNX] Fix reciprocal when input is not floating point (#67471) (#67808)

garymm committed 2 years ago

f57c6303

[ONNX] Use human readable enum for dtype scalars (#66822) (#67807)

garymm committed 2 years ago

eb22d06e

[ONNX] Fix new_full and full_like for Python 3.9 (#67124) (#67806)

garymm committed 2 years ago

958d5176

[ONNX] Support opset 15 (#67121) (#67805)

garymm committed 2 years ago

37688148

[ONNX] Suppress ort warnings in onnx related test (#67054) (#67804)

BowenBao committed 2 years ago

ead59b5f

Op info for activation functions 2 (softsign, tanh, tanhshrink, threshold, celu, sigmoid, mish, hardsigmoid) (#67492)

Samantha Andow committed 2 years ago

ea60e7d5

Avoid convert trt.Dims to tuple in hot path (#67960)

Yinghai Lu committed 2 years ago

a1d733ae

[Quant] Add dynamic QAT Linear module (#67325)

andrewor committed 2 years ago

4a8f2744

`torch.lobpcg.backward`: do not save non-Variable types with `ctx.save_for_backward`. (#67994)

nikitaved committed 2 years ago

db456d16

[lint] small updates to .lintrunner.toml (#67942)

suo committed 2 years ago

8e252813

[lint] Add the rest of the grep linters (#67932)

suo committed 2 years ago

d201102d

Newer Older