strisunshinewentingwang/pytorch

Pull Requests Commits

Replace AutoNonVariableTypeMode with InferenceMode in fbcode. (#55114)

Ailing Zhang committed 3 years ago

24c90495

Split copy_ kernel to InplaceOrView. (#55133)

Ailing Zhang committed 3 years ago

181de406

Don't globally disable any ShellCheck warnings (#55165)

samestep committed 3 years ago

09670c7d

Revert D25399470: add channels last for MaxPool2d

agolynski committed 3 years ago

978fca64

Modified lstsq_helper to accept lapack error codes tensor (#54720)

IvanYashchuk committed 3 years ago

e406d4e6

ns for fx: weight extaction for conv1d and conv3d (#55079)

vkuzo committed 3 years ago

8062545c

ns for fx: ensure kwargs are handled when graph matching (#55078)

vkuzo committed 3 years ago

80b1b7e4

ns for fx: clean up debug print statements (#55077)

vkuzo committed 3 years ago

a590fa7a

ns for fx: move it to top level file (#55060)

vkuzo committed 3 years ago

f6b25e75

ns for fx: weight extraction for nni.ConvReLU2d (#54335)

vkuzo committed 3 years ago

c6cb99a6

ns for fx: make input logging work for multi node subgraphs (#54327)

vkuzo committed 3 years ago

5319d17b

ns for fx: make input logging work for multi-node subgraphs (#54326)

vkuzo committed 3 years ago

b8019cee

ns for fx: refactor test cases (#54280)

vkuzo committed 3 years ago

cbcde790

ns for fx: add support for shadowing linear fp16 patterns (#54275)

vkuzo committed 3 years ago

757e3cbf

add channels last for MaxPool2d (#48917)

mingfeima committed 3 years ago

f43eb59a

Revert D27478436: Use tensorpipe::Buffer::device() instead of tensorpipe::Buffer::deviceType().

Mike Ruberry committed 3 years ago

bdb225e9

Use tensorpipe::Buffer::device() instead of tensorpipe::Buffer::deviceType().

Lucas Hosseini committed 3 years ago

3e185253

[ATen][qembeddingbag] Avoid tensor refcount bumps (#55023)

Hao Lu committed 3 years ago

61914cb2

[c10] Add default constructor to Maybeowned (#55128)

Hao Lu committed 3 years ago

93d0f636

Adds torch.* API section for TorchScript Lang Ref (#53236)

gmagogsfm committed 3 years ago

ec609e74

[PyTorch Edge] Provide a method ObservedOperators::getUnobservedOperatorList() so that model tracer can empty it out during tracing (#55017)

dhruvbird committed 3 years ago

271879fe

Transition to new tensorpipe::Pipe API. (#55193)

Lucas Hosseini committed 3 years ago

09f1f145

Port torch.copysign method_tests() to OpInfo (#54945)

RockingJavaBean committed 3 years ago

b074a243

[OpInfo] Enable jit tests for multi_dot (#55147)

heitorschueroff committed 3 years ago

ed4a1d54

[TensorExpr] Add TORCH_APIs to all expr classes. (#55002)

Mikhail Zolotukhin committed 3 years ago

ff6b3c76

[TensorExpr] Cleanup IRNodeType enum. (#55001)

Mikhail Zolotukhin committed 3 years ago

1ccaec02

[TensorExpr] Remove stale docs from DesignOverview.md. (#55000)

Mikhail Zolotukhin committed 3 years ago

f8f30a5e

[TensorExpr] Nuke BaseCallNode. (#54999)

Mikhail Zolotukhin committed 3 years ago

bdbfb2a0

[TensorExpr] Nuke FunctionCall. (#54998)

Mikhail Zolotukhin committed 3 years ago

0b75f862

[TensorExpr] Nuke DepTracker and findAllNeededTensors. (#54997)

Mikhail Zolotukhin committed 3 years ago

688e3507

Newer Older