strisunshinewentingwang/pytorch

Pull Requests Commits

Use the same checks in all `grid_sampler` functions

Nikita Karetnikov committed 2 years ago

936a6505

Dedupe no parsing __torch_function__ handler

ezyang committed 2 years ago

e3848d75

Introduce SafePyObject, make TorchDispatchTypeObject use it

ezyang committed 2 years ago

de6353ba

Split PyInterpreter into its own file.

ezyang committed 2 years ago

1faf1cdf

[Quant][fx] Define native backend_config_dict for linear and conv (#74636)

andrewor14 committed 2 years ago

ee9335a6

Revert D35352705: fix: buck2 build fbcode//caffe2:torch_types_gen

ezyang committed 2 years ago

7b506e88

fix: buck2 build fbcode//caffe2:torch_types_gen (#75176)

luciang committed 2 years ago

152489a8

Add BFloat16 support for smooth_l1_loss on CPU (#62558)

CaoE committed 2 years ago

c5872e6d

Deprecate bytecode v3 and bump kMinSupportedBytecodeVersion to 4 (#75149)

cccclai committed 2 years ago

96050ee0

Restore TestTorchFunctionOverride

peterbell10 committed 2 years ago

bf165526

torch.tensor: add tests for list of numpy arrays case

kshitij12345 committed 2 years ago

05090224

Update forward AD not supported error message

soulitzer committed 2 years ago

631f0351

WIP Jiterator reduction

ngimel committed 2 years ago

ad028e5e

[quant][fx] Remove is_output_quantized from QuantizeHandler (#74843)

jerryzh168 committed 2 years ago

bd032cd8

[WIP] Add support to `Tensor[]` for structured kernel codegen.

ezyang committed 2 years ago

dca42dc9

Skip TorchScript backend for OVRSource as well (#75138)

malfet committed 2 years ago

8b7e2bf7

Save disable_torch_function in ThreadLocalState

ezyang committed 2 years ago

3f108a5c

Rewrite upgrader bytecode version from 3 to 4 (content unchanged) (#75120)

cccclai committed 2 years ago

6efc5c1a

Extend jit::load to work on flatbuffer file (#75022)

pavithranrao committed 2 years ago

d9d34922

[Easy][FSDP] Update full osd warning (#75109)

awgu committed 2 years ago

3ebdd438

Dynamo+LTC: merging related code from staging branch to master (#75046)

shunting314 committed 2 years ago

19747cbb

[PyTorch] Flip polarity of masked_softmax mask (#78)

swolchok committed 2 years ago

8f4f1638

Fix SyncBatchNorm for empty inputs (#74944)

mrshenli committed 2 years ago

87ab665b

[JIT] OpInfo tests for nvfuser (#71299)

davidberard98 committed 2 years ago

c5b3727e

[JIT] Enable NVFuser tests in OSS CI (#73322)

davidberard98 committed 2 years ago

27deefb5

[JIT] Optionally validate nvfuser outputs after execution (#74361)

davidberard98 committed 2 years ago

e9e75215

Revert "Nvfuser guard patch"

pytorchmergebot committed 2 years ago

1352c641

[ROCM] unskip test_fn_grad

micmelesse committed 2 years ago

637273da

[GHA] Do not chown for linux CI

malfet committed 2 years ago

623f9397

[ROCm] re-enable test_Conv2d_groups_nobias tests

KyleCZH committed 2 years ago

f888dc58

Newer Older