strisunshinewentingwang/pytorch

Pull Requests Commits

[Quant] Add out variant for int8 quantized::linear (#58282)

Hao Lu committed 3 years ago

4d7abdbd

[nnc] Enable CPU fusion inside Facebook, take 2 (#58347)

bertmaher committed 3 years ago

c76405d3

VaryingShape<Strides>::isComplete() needs to consider whether each Stride is complete (#58510)

bertmaher committed 3 years ago

dcfc2050

[nnc] Do not fuse unsqueeze with variable dim (#58346)

bertmaher committed 3 years ago

3d20ddfe

[nnc] Make the pretty printer prettier (#57874)

bertmaher committed 3 years ago

2ddd8416

[jit] Add a utility class SourceRef to represent Source as keys (#57396)

zhxchen17 committed 3 years ago

3a3959d2

[BE] Use __func__ as checkAllSameGPU() 1st arg (#58502)

malfet committed 3 years ago

0362b753

move unused parameters to end of bucket orders when rebuild buckets for static graph (#58097)

zhaojuanmao committed 3 years ago

ea0f7c47

[PyTorch Edge] bytecode version bump to v5 and enable share constant table (#57888)

cccclai committed 3 years ago

a7b62abe

[nnc][scripts] Add a script for bisecting the TE fuser pass (#58357)

bertmaher committed 3 years ago

9eee782c

removing old comment (#56430)

Krovatkin committed 3 years ago

7d78d72d

Comment why render_test_results is its own step (#58505)

samestep committed 3 years ago

a07cd22e

Add long tensor type to AddFakeFp16 Op (#58504)

venkatacrc committed 3 years ago

8efaab1b

[NNC] Do not optimize conditionals when the corresponding loop is not normalized (#57675)

navahgar committed 3 years ago

4b859cbc

[NNC] Add a method to check if a loop is normalized (#57674)

navahgar committed 3 years ago

a71b99b5

[NNC] Optimize conditionals that correspond to the form generated for aten::cat op. (#57673)

navahgar committed 3 years ago

3fe72d30

[Pytorch Sparsity] Add sparse sources to build target

kimishpatel committed 3 years ago

db42ec42

Support symbolic diff for leaky_relu (#58337)

Edvard Ghazaryan committed 3 years ago

ad97fd80

Clarify .github/scripts/generate_ci_workflows.py (#58498)

samestep committed 3 years ago

e1551f16

[PyTorch] Add a guard rail to TensorIterator::add_borrowed_{in,out}put (#58279)

swolchok committed 3 years ago

5fcf49f5

[PyTorch] Migrate remaining CUDA TI usage to borrowing where possible (#58278)

swolchok committed 3 years ago

03f2f0f8

[PyTorch] Migrate CUDA indexing TI usage to borrowing (#58277)

swolchok committed 3 years ago

1fd256dc

[PyTorch] Migrate TensorAdvancedIndexing TI usage to borrowing where possible (#58276)

swolchok committed 3 years ago

029289bd

[PyTorch] Migrate all extant uses of build_binary_float_op to build_borrowing_binary_float_op (#58273)

swolchok committed 3 years ago

439ba27d

[PyTorch] Migrate all extant uses of build_binary_op to build_borrowing_binary_op (#58272)

swolchok committed 3 years ago

8a4a511f

Fix KeyError returned by _maybe_get_last_node_only_observer (#58443)

mleshen committed 3 years ago

07da584d

Simplify .github/scripts/generate_ci_workflows.py (#58491)

samestep committed 3 years ago

46484e8d

Collect kernel version (#58485)

malfet committed 3 years ago

f7c15610

fix nonzero perf regression (#58468)

ngimel committed 3 years ago

92e36240

[local lint] Remove success checks in tests (#58490)

driazati committed 3 years ago

4ce8378e

Newer Older