strisunshinewentingwang/pytorch

Pull Requests Commits

Revert D29190420: [nnc][tests] Tests and benchmarks for computeSum

anjali411 committed 3 years ago

b14f19b6

To add edge_order=2 and documentation for gradient operator (#58165)

iramazanli committed 3 years ago

90cd57ee

[NormalizeArgs] Retain node.meta (#60449)

jfix71 committed 3 years ago

7ed07e2a

Ensure num_threads is initialized before calling omp_get_max_threads (#60185)

peterbell10 committed 3 years ago

66452e0a

OpenMP: Refactor parallel_reduce to share code with parallel_for (#60184)

peterbell10 committed 3 years ago

19553438

Ensure thread id is valid in nested parallel regions (#60183)

peterbell10 committed 3 years ago

c75714e5

Migrate crossKernel from THC to ATen (CUDA) (#60039)

peterbell10 committed 3 years ago

3f3fd570

[BE] Fix Convolution.cpp build warnings (#60463)

malfet committed 3 years ago

f590ccea

Increase tolerance for test_grad_scaling_clipping (#60458)

Flamefire committed 3 years ago

3846cef2

`topk` on CUDA supports `bfloat16` (#59977)

eqy committed 3 years ago

40de03fc

[nnc][tests] Tests and benchmarks for computeSum (#60160)

bertmaher committed 3 years ago

21479ad2

[nnc] Speed up batchnorm benchmark

bertmaher committed 3 years ago

fbeb8b49

[pytorch][nnc] external function call to xnnpack ops (#59525)

ljk53 committed 3 years ago

b0c9762e

Add error message for sequence length to be equal to 0 case for RNNs (#60269)

iramazanli committed 3 years ago

79dc500a

Add custom code filter for TS (#60309)

nikithamalgi committed 3 years ago

dc9aa7b9

[quant] Input-Weight Equaliaztion - convert modifications (#59963)

angelayi committed 3 years ago

3de79b77

Enable rcb lookup for typing (#60413)

nikithamalgi committed 3 years ago

7589d9c5

avoid unnecessary copies in MultiDispatchKeySet (#60093)

nunoplopes committed 3 years ago

135e203e

[quant] avoid resize calls in observer/fake_quant (#60386)

supriyar committed 3 years ago

4887c6e4

parse_reports() should include hidden files (#60404)

malfet committed 3 years ago

d3ae3e07

Remove some unused variables (#60411)

r-barnes committed 3 years ago

986a8805

Fix some variable types (#60414)

r-barnes committed 3 years ago

36d4062a

Fix some loop types (#60415)

r-barnes committed 3 years ago

7d779f84

Fix lint (#60472)

samestep committed 3 years ago

6e926f13

up the priority of numpy array comparisons in self.assertEqual (#59067)

pmeier committed 3 years ago

0c916c8a

Do not wrap Tensor.{grad,_base} by default (#60464)

ezyang committed 3 years ago

82c52fd4

Disable warn_unused_ignores again (#60480)

samestep committed 3 years ago

f42140cb

Implement erfcx() (#58194)

wqwu-fb committed 3 years ago

6a87e8d0

Improve testing of inplace views (#59891)

soulitzer committed 3 years ago

b3496543

[caffe2/utils] Add explicit rule to avoid package boundary violation

andrewjcg committed 3 years ago

20bda005

Newer Older