strisunshinewentingwang/pytorch

Pull Requests Commits

Reland: "free up dispatch key space (in C++)" (#74963)

bdhirsh committed 2 years ago

1b7d7d93

Revert "Removed python dispatch keys from dispatch key extraction"

pytorchmergebot committed 2 years ago

3d34afa9

Adding versions to flatbuffer schema (#74989)

pavithranrao committed 2 years ago

6905feea

expanded weights: group norm faster rule

samdow committed 2 years ago

72f7193f

[PyTorch Edge] Add Quantized Softmax Op (Naive Implementation) (#75017)

salilsdesai committed 2 years ago

8d7242a1

expanded weights: layer norm faster rule

samdow committed 2 years ago

8b8f3e83

[ROCm] unskip FFT tests

micmelesse committed 2 years ago

c4321e13

remove unused nn_path from generate_code (#74563)

mikey dagitses committed 2 years ago

60729d02

Use ONNX Runtime 1.11 in CI

garymm committed 2 years ago

1a1f0d4b

Revert D35000703: [WIP][FSDP] Mixed precision enablement

malfet committed 2 years ago

a98d1a5f

Catch overflows in calculating storage byte size

peterbell10 committed 2 years ago

13a3e5c7

use the shared //tools/codegen:gen in OSS Bazel (#74471)

mikey dagitses committed 2 years ago

40bf3cfe

move codegen binary to the common build system (#74470)

mikey dagitses committed 2 years ago

785972b4

Add Hpu to the rebuild component list

aymanhabana committed 2 years ago

88096253

[WIP][FSDP] Mixed precision enablement (#74452)

rohan-varma committed 2 years ago

6b0b088c

use the //tools/codegen target in Bazel (#74465)

mikey dagitses committed 2 years ago

79307fbd

move //tools/codegen:codegen into shared build structure (#74386)

mikey dagitses committed 2 years ago

76eabe9e

Make all `.pyi.in` files exportable from torch/_C/ folder (#74962)

malfet committed 2 years ago

a78eb243

Revert "Reland "[pytorch][PR] Support dataclasses in TorchScript" take 2 (#74353)"

malfet committed 2 years ago

fa1a41ca

Revert D35258695: [quant][fx] Cleanup unused to_fp16 check code in lowering

malfet committed 2 years ago

0b845bb6

Nvfuser code bump 030122 (#73627)

jjsjann123 committed 2 years ago

873ced7c

[quant] Fix qmin/qmax when using customized qrange (#74717)

terrychenism committed 2 years ago

b82df92c

[Deploy] Change `numModules` type to `unsigned` (#74978)

malfet committed 2 years ago

ce700da7

[quant] fix int16 quantization scale in conv weight (#74665)

terrychenism committed 2 years ago

8c05f44f

Revert D35194935: Check all CUDA API calls for errors in torch/

malfet committed 2 years ago

c40a009d

[quant][ns] Fix ns tool bug for mobilenetv2/v3 (#74149)

terrychenism committed 2 years ago

3c701468

Check all CUDA API calls for errors in torch/ (#74923)

r-barnes committed 2 years ago

79e5b053

Revert D34647822: [tensorexpr] Add support for aten::stack

malfet committed 2 years ago

43313cbd

Revert D34808051: [tensorexpr] Enabled aten::stack in the fuser pass with static shapes

malfet committed 2 years ago

320e5a82

[quant][fx] Cleanup unused to_fp16 check code in lowering (#74969)

jerryzh168 committed 2 years ago

ec6f7670

Newer Older