strisunshinewentingwang/pytorch

Pull Requests Commits

[ROCm] persist test results even on failure

jeffdaily committed 2 years ago

177ea463

[ROCm][GHA] keep docker images for at most 1 day

jeffdaily committed 2 years ago

6d1bacaf

Port to_padded_tensor CUDA kernel from pytorch/nestedtensor

cpuhrsch committed 2 years ago

ea190169

[PyTorch] Use native serial stack when there is only 1 thread (#76399)

Mike Iovine committed 2 years ago

1721137f

Support for tensor subclasses as parameters

jbschlosser committed 2 years ago

bc34cf5f

Add "mps" device to PyTorch framework.

kulinseth committed 2 years ago

54c75e1e

Add new dispatch keys for Fake Tensor and Deferred Module Initialization

cbalioglu committed 2 years ago

a0bf0f56

Revert D35860266: [pytorch][PR] Update torch::lazy::BackendDevice to have a new default ordinal

Jiewen Tan committed 2 years ago

a28b132b

[ROCm] Skipping few multiprocess test

pruthvistony committed 2 years ago

2469525c

Enable LTC Input/Output Mapping (#75828)

henrytwo committed 2 years ago

a39a2c39

[SR] Fix quantized linear tests not managing outputs (#75776)

Mike Iovine committed 2 years ago

3fa77fa5

remove unneeded overload for nansum

ngimel committed 2 years ago

04b33133

Revert D35511873: [iOS][coreml] Add CoreML memory observer

janeyx99 committed 2 years ago

12299991

[primTorch] Prototype tracer and elementwise unary reference opinfo class

Mike Ruberry committed 2 years ago

4048d4cd

fix nested grad(functionalize(f)) transforms

bdhirsh committed 2 years ago

aae7b00f

Fix typehint of multi_head_attention_forward

ultmaster committed 2 years ago

3ac27e78

[quant][core][gpu][improvement] Integrated quantized cudnn max pool2d with existing quantized_max_pool2d (#76129)

dzdang committed 2 years ago

6e292f1a

add buck generated files to ignore list

linbinyu committed 2 years ago

6e959dec

rebase via comment

clee2000 committed 2 years ago

0bd8e237

[fx2trt] support ops for hf_T5 (#62)

Wei Wei committed 2 years ago

f323a8ab

ci: Unblock syncbranches, add a58c6ae and 7106d21 to block list (#76417)

seemethere committed 2 years ago

48ea440b

[caffe2] drop XROS ports (#76366)

h-friederich committed 2 years ago

5932c371

[iOS][coreml] Add CoreML memory observer (#76251)

SS-JIA committed 2 years ago

22fb9294

[libkineto] Re-enable user-annotations in PyTorch (#75601)

aaronenyeshi committed 2 years ago

09a5b075

Disable RPC profiling for kineto profilers

rohan-varma committed 2 years ago

ec62901a

Add mixed precision doc

rohan-varma committed 2 years ago

5cd880f4

Reenable XLA workflow/test

yeounoh committed 2 years ago

3c327f3f

Fix TensorImpl use count assert for _thnn_fused_lstm_cell_backward

soulitzer committed 2 years ago

6eee1e9b

[JIT] Register decomp reland

Elias Ellison committed 2 years ago

81b9cb74

[mobile test] Update and rename README to README.md

linbinyu committed 2 years ago

4d1d1b31

Newer Older