strisunshinewentingwang/pytorch

Pull Requests Commits

remove index_fill side-effect for scalar tensors (#52209)

nikitaved committed 3 years ago

0048d97e

[TensorExpr] Add `Placeholder::handle` method to get the corresponding `BufHandle`. (#52793)

Mikhail Zolotukhin committed 3 years ago

57947c5d

[TensorExpr] Add an unmasked `Load` constructor. (#52790)

Mikhail Zolotukhin committed 3 years ago

d3b427a0

Introduce `mlc` device (ML Compute device) to PyTorch's device list (#50634)

bmon581 committed 3 years ago

30cb6ac5

Drop unused variables (#52643)

r-barnes committed 3 years ago

2bdf6305

Added fast path in the case of no hooks (#52576)

Chillee committed 3 years ago

a649d808

Add 64bit indexing support for softmax (#52713)

zasdfgbnm committed 3 years ago

a6b7da7d

Use finer-grained mutexes in TensorPipe RPC agent (#52749)

lw committed 3 years ago

c140a5ec

print matrix dims in torch cuda matrix multiply error (#52780)

ajaech committed 3 years ago

c9548176

Use c10::irange for great good (#52153)

r-barnes committed 3 years ago

29c4290a

Modernize for-loops in caffe2/torch (#52618)

r-barnes committed 3 years ago

373a20ad

Kernel launch checks for aten/src/ATen (#52185)

r-barnes committed 3 years ago

0567988e

Update Gloo submodule (#52754)

lw committed 3 years ago

98873b92

Implemented torch.linalg.multi_dot (#51807)

heitorschueroff committed 3 years ago

0396f492

Add torch.linalg to generated annotated_args for test_overrides (#52464)

heitorschueroff committed 3 years ago

964d47df

[quant] Reference option for conv module (#52316)

jerryzh168 committed 3 years ago

7b54a8fc

[Pytorch Mobile] Improve Bundled Inputs Error Checking (#52386)

JacobSzwejbka committed 3 years ago

3cf08eaf

[TensorExpr] LoopNest: Cleanup LoopNest constructors. (#52726)

Mikhail Zolotukhin committed 3 years ago

88a160dc

structured kernels - error check when structured_delegate is not marked structured (#52227)

bdhirsh committed 3 years ago

08d26694

Catch Flake8 error codes with multiple letters (#52750)

samestep committed 3 years ago

5e977d9c

[ZeroRedundancyOptimizer] Remove pseudo futures handling, not needed (#52698)

blefaudeux committed 3 years ago

7ae77686

Clarify usage and output of tools/test_history.py (#52640)

samestep committed 3 years ago

27d04f29

[FX acc]Add fx_glow support for multi outputs (#52527)

842974287 committed 3 years ago

b4b7db2f

Change `maybe_resize_storage_cpu` `new_size` arg to unsigned (#52671)

malfet committed 3 years ago

59ac0ff0

Add discontiguous kwarg to make_tensor (#51985)

heitorschueroff committed 3 years ago

08d7f296

Fix the ordering of TCPStore's compare_set parameters (#52696)

cbalioglu committed 3 years ago

3489b4a7

[Reland] Update XNNPACK (#52691)

malfet committed 3 years ago

97b6b3df

[nnc] Benchmarks for concat (#52592)

navahgar committed 3 years ago

8af64835

Revert D26599390: [pytorch][PR] Fix for incorrect usage of logging in torch/distributed/distributed_c10d.py

H-Huang committed 3 years ago

b56f59ea

[fx/package] make GraphModules packageable (#51976)

suo committed 3 years ago

958d9a83

Newer Older