strisunshinewentingwang/pytorch

Pull Requests Commits

Catherine Lee committed 3 years ago

9d745be7

Catherine Lee committed 3 years ago

e37491c8

Catherine Lee committed 3 years ago

9dbc2618

Catherine Lee committed 3 years ago

54b5867a

Catherine Lee committed 3 years ago

82807ef2

print xpass and fail

Catherine Lee committed 3 years ago

e0c4fa64

[Dynamo] Fix str(Guard.obj_weakref) bug to re-ennable support overriding __getattr__ (#88564)

yanboliang committed 3 years ago

6fe47b68

[DataLoader] Removing DataLoader2 related code (#88848)

NivekT committed 3 years ago

be8d88f8

Make InductorCPU usable in internally (#88870)

malfet committed 3 years ago

f39cad50

[ONNX] Pretty print diagnostic logging (#88261)

BowenBao committed 3 years ago

fbc18782

[tourch] BatchBoxCox - fix numerical issue in vectorized code (#88875)

efiks committed 3 years ago

ea0ec9d7

Fix unused variable 'options' warning in RNN.cpp (#88753)

r-barnes committed 3 years ago

dfb4b73e

[FSDP][state_dict][5/N] Remove the FSDP module dependency from _state_dict_utils (#88637)

fegin committed 3 years ago

7aa144ac

Fix CUDNN_PATH handling on Windows (#88898)

malfet committed 3 years ago

575e02df

[fix] allow saving python attr on Tensor and Parameter via torch.save (#81616)

kshitij12345 committed 3 years ago

f7494632

Revert "rename DisableTorchFunction to DisableTorchFunctionSubclass (#88218)"

pytorchmergebot committed 3 years ago

ba4d5aae

Revert "add DisableTorchFunction that matches DisableTorchDispatch (#88219)"

pytorchmergebot committed 3 years ago

4e5d7afe

[ONNX] Add stack info to diagnostics (#87258)

BowenBao committed 3 years ago

9d7d21f5

[FSDP][state_dict][4/N] Move the core logic of summon full parameters to _unshard_params_utils.py (#88636)

fegin committed 3 years ago

3d1c5c89

Fix ATen Fallback for BUILD_CAFFE2=0 for ONNX-only ops (#88504)

Thiago Crepaldi committed 3 years ago

5f0783bd

Take input striding for conv forward based on eager output (#88706)

eellison committed 3 years ago

8ff2e34c

Revert "[Autograd] Use in-place input accumulation fast path for dense Tensors. (#88339)"

pytorchmergebot committed 3 years ago

adfbd831

Explicitly check filelike arg of `torch.save` (#88867)

kurtamohler committed 3 years ago

89a326ff

Regularize bernouilli_ with bernouilli decomp (#88349)

eellison committed 3 years ago

a6832b08

Symintify `broadcast_to` (#88776)

nkaretnikov committed 3 years ago

1e8f95ac

Add meta impl for topk (#88694)

anjali411 committed 3 years ago

d615d122

[FSDP][state_dict][3/N] Change how state_dict utils access attributes in _FSDPState (#88635)

fegin committed 3 years ago

3c7f9666

Add context manager to allow mutation on saved tensors (#79056)

soulitzer committed 3 years ago

b92acee8

[dynamo] Add torch.device to is_safe_constant (#88766)

desertfire committed 3 years ago

91b71cdb

[FSDP][state_dict][2/N] Move state_dict related enums/dataclasses/states to state_dict_utils.py, api.py and init_state_dict() (#88481)

fegin committed 3 years ago

324ac93a

Older