strisunshinewentingwang/pytorch

Pull Requests Commits

Update TensorPipe submodule (#52677)

lw committed 3 years ago

3ff6c917

Add scatter_add to amp promote list (#52133)

pbialecki committed 3 years ago

39fa0b5d

fix(docs): remove redundant hardsigmoid() in docstring to show up `inplace` parameter (#52559)

Jeff Yang committed 3 years ago

316eabe9

ns for fx: update graph matching to handle dicts and tuples in node args (#52681)

vkuzo committed 3 years ago

1618dc2a

ns for fx: update graph matching to not match nodes with equal types (#52402)

vkuzo committed 3 years ago

608f44b2

ns for fx: support linear_relu for weight matching (#52395)

vkuzo committed 3 years ago

4483c48e

ns for fx: allow graph matching of parents of cat (#52368)

vkuzo committed 3 years ago

64b4e37c

[Pytorch, sparsity] Bug fix to update requantization and zp parameters of input (#52797)

kimishpatel committed 3 years ago

13121598

Lower ReLu6 to aten (#52723)

bmon581 committed 3 years ago

99a428ab

Update backwards compatibility check to ignore reverted op (#52841)

lw committed 3 years ago

fa7575ea

Fix typos in tools/test_history.py helpstring (#52840)

samestep committed 3 years ago

91412690

Fix RPC get_worker_info for rank=0 (#52804)

mrshenli committed 3 years ago

1ac59d9d

Store test filename in test report path (#52791)

janeyx99 committed 3 years ago

f71d9e28

Revert D26375734: Implemented torch.linalg.multi_dot

lw committed 3 years ago

92a4ee1c

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

Newer Older