strisunshinewentingwang/pytorch

Pull Requests Commits

[BE][Docs] Improve dist.new_group doc (#55660)

rohan-varma committed 3 years ago

19f15317

docs: improve torch.matrix_exp() (#55626)

neal2018 committed 3 years ago

a3c062d4

Remove legacy constructor calls from pytorch codebase. (#54142)

ysiraichi committed 3 years ago

93bf0ae6

[JIT] Allow unpacking tuple and assign their values to SELECT-type expressions (#55268)

gmagogsfm committed 3 years ago

fa29a647

Disambiguate error message for working with not fully refined tuple types (#55745)

tugsbayasgalan committed 3 years ago

b80c6f86

Make leak_corrupted_threadpool non-atomic (#55341)

imaginary-person committed 3 years ago

facbcec2

Optimize constructing tensors from external data (#55705)

cbalioglu committed 3 years ago

84a7ab25

torch.testing allclose -> close (#54781)

pmeier committed 3 years ago

255494c2

change torch.testing helper asserts to checks (#54780)

pmeier committed 3 years ago

c9b94a85

[PyTorch] Add & use inferExpandGeometry_dimvector (#55316)

swolchok committed 3 years ago

548765d9

[PyTorch][easy] Use sizes()[x] instead of size(x) in addr (#55247)

swolchok committed 3 years ago

151869ac

[PyTorch] Update expand_size API to match expand_inplace (#55246)

swolchok committed 3 years ago

12c19c39

[PyTorch] Update expand_outplace API to match expand_inplace (#55245)

swolchok committed 3 years ago

16a9141e

[PyTorch] Add MaybeOwned::operator*() && (#55244)

swolchok committed 3 years ago

6fd87592

[PyTorch] Use infer_size_dimvector in ExpandUtils (#55180)

swolchok committed 3 years ago

e8dd6510

[PyTorch] New expand_inplace API with MaybeOwned<Tensor> and no unary tuples (#55065)

swolchok committed 3 years ago

fa19b6dd

[Gradient Compression] Fix PowerSGD docstring by removing an extra whitespace (#55666)

Yi Wang committed 3 years ago

2496a093

Revert D27691509: Replace AutoNonVariableTypeMode with InferenceMode in static runtime.

Ailing Zhang committed 3 years ago

5a8cdc2f

Replace AutoNonVariableTypeMode with InferenceMode in static runtime.

Ailing Zhang committed 3 years ago

d695ba94

[tensorexpr] Add PYTORCH_TENSOREXPR_DONT_FUSE env variable to disable fusion on specified operators - fixed #50757 (#55650)

huiguoo committed 3 years ago

263a15c5

[reland][quant][graphmode][fx][refactor] Factor out insert_observers_for_model to a separate function (#54733) (#55307)

jerryzh168 committed 3 years ago

3e8ebb17

Fix type annotations for state_dict() override (#55704)

malfet committed 3 years ago

d33829f8

OpInfo for kron (#55546)

asuhan committed 3 years ago

fc349cbc

[SPMD] Remove the code branches only used in SPMD mode from distributed.py (#55353)

Yi Wang committed 3 years ago

3e9cbe5e

[Hackathon] Add source highlighting check to test_unsupported_ops (#55501)

gmagogsfm committed 3 years ago

717d54bc

Revert D27670883: [pytorch][PR] Added an OpInfo for mm & ported its method_tests

kimishpatel committed 3 years ago

7485818a

mark embedding backward non-deterministic for max mode rather than all reducing modes (#55574)

yuguo68 committed 3 years ago

846c8d94

Make VariableVersion::DISABLED the default constructor for VariableVersion. (#55572)

Ailing Zhang committed 3 years ago

7671c15d

Fix annotations in _autograd.pyi (#55706)

malfet committed 3 years ago

6e4e3a11

[android] Module load extraFiles (#55644)

IvanKobzarev committed 3 years ago

ee2de8ae

Newer Older