strisunshinewentingwang/pytorch

Pull Requests Commits

Declare public dependencies on libshm (#82694)

peterbell10 committed 2 years ago

5a8b07de

Merge more symbolic meta kernels and symint changes from branch (#86334)

bdhirsh committed 2 years ago

08e3999f

Release 1.13: Bump nightly version 1.13->1.14 (#86296)

atalman committed 2 years ago

3af0eafe

Fix SparseAdam consuming iterator (#86210)

ssnl committed 2 years ago

5ed75ec1

[FSDP] Doc to explain running submodules (#86343)

rohan-varma committed 2 years ago

f0977c46

[FSDP] Fix clip_grad_norm for CPU offload (#86337)

rohan-varma committed 2 years ago

3db8ddca

[FSDP] assert to runtime error (#86336)

rohan-varma committed 2 years ago

adfd8f38

CheckpointSequential support non-reentrant (#86331)

rohan-varma committed 2 years ago

7a411952

Docs: fix typo (#86273)

cherrywoods committed 2 years ago

3037f3d7

Revert "Fix memory leak in _LRScheduler.step() (#85602)"

pytorchmergebot committed 2 years ago

233d6f19

Fix for binary upload step, use bash shell rather then default sh (#86382)

atalman committed 2 years ago

bf746798

[ao] fixing public v private for qconfig.py (#86026)

HDCharles committed 2 years ago

facf210f

[kineto] guard global observer init against Edge profiler (#86347)

chaekit committed 2 years ago

7c5e07f8

[torch.ao.quantization] include torch.qint32 for static quant (#86345)

jiaxuzhu92 committed 2 years ago

bc919ac7

Two small improvements to references (#86371)

lezcano committed 2 years ago

08780229

Add total GPU memory utilization (#86250)

huydhn committed 2 years ago

795906f2

Make mergebot message clearer when starting a new merge (#86311)

ZainRizvi committed 2 years ago

1059d3b5

Enable autograd on Linear with sparse COO weight (#86302)

pearu committed 2 years ago

6b295cd0

Support autograd on sparse_mm in full. (#86301)

pearu committed 2 years ago

8f2c2167

Support sum on a sparse COO tensor. (#86300)

pearu committed 2 years ago

88b882cd

Support autograd on Linear with sparse compressed weight. (#86137)

pearu committed 2 years ago

f104490d

Enable sparse_dim() and dense_dim() methods for Strided tensors (#86203)

pearu committed 2 years ago

fc21cc82

[torchdynamo hash update] update the pinned torchdynamo hash (#86306)

pytorchmergebot committed 2 years ago

bed1ece9

Fix memory leak in _LRScheduler.step() (#85602)

KinglittleQ committed 2 years ago

eb32330d

Ensure the minimum NVIDIA driver version to be 515.57 for CUDA 11.7 (#86344)

huydhn committed 2 years ago

b8b564c9

Remove extra bracket, update header definition (#86317)

cpuhrsch committed 2 years ago

0c148a4b

Use FindCUDAToolkit to find cuda dependencies (#82695)

peterbell10 committed 2 years ago

fb9b9659

[MPS] Better error message for `slow_conv2d_forward` (#86303)

malfet committed 2 years ago

fa799132

Inline asIntArrayRef (#86350)

ezyang committed 2 years ago

4d772889

[Quant] Remove weight from DTypeConfig for non-weighted ops (#86335)

andrewor14 committed 2 years ago

cebf08af

Newer Older