strisunshinewentingwang/pytorch

Pull Requests Commits

BFP16 quantization/dequantization (#62974)

marjanfariborz committed 3 years ago

5a2d4c1f

Migrate Embedding thrust sort to cub sort (#62495)

zasdfgbnm committed 3 years ago

a2c4b18b

Use output memory format based on input for cudnn_convolution_relu (#62482)

= committed 3 years ago

0e45f99e

irange-ify 12 (#62120)

r-barnes committed 3 years ago

9f0cfc2d

irange-ify 1 (#62193)

r-barnes committed 3 years ago

6d07387d

Fix render_test_results if condition on always() (#62997)

zhouzhuojie committed 3 years ago

91db9cdc

[reland] Gate DistributedOptimizers on RPC availability (#62937)

rohan-varma committed 3 years ago

2144b43a

irange-ify 8d (#62505)

r-barnes committed 3 years ago

c6b669d3

[fx] store Tracer class on Graph and GraphModule for package deserialization (#62497)

bradleyhd committed 3 years ago

2fc9f15a

Mark unused functions with `C10_UNUSED` (#62929)

malfet committed 3 years ago

d67963a2

Stop exporting symbols in anonymous namespaces (#62952)

peterjc123 committed 3 years ago

9f85712f

[Static Runtime] Add tests for all aten ops (#62347)

Mike Iovine committed 3 years ago

779c09fa

handle get_attr opearations in typechecker (#62682)

migeed-z committed 3 years ago

15ec80fb

Linker version script to hide LLVM symbols (#62906)

bertmaher committed 3 years ago

b9f2875b

Add ``allow_empty_param_list`` to functional optimizers (#62522)

awgu committed 3 years ago

afe9d82d

[Vulkan] Added Hardshrink op (#62870)

beback4u committed 3 years ago

7df624f3

Change output node handling for typechecker to deal with tuples (#62582)

migeed-z committed 3 years ago

28274510

[LTC] Improve QUICKSTART.md and OP_LOWERING_GUIDE.md (#63123)

Jiewen Tan committed 3 years ago

Verified 5fe2f262

[LTC] Add _log_softmax TorchScript lowering (#63105)

Jiewen Tan committed 3 years ago

Verified 53833fb8

Merge branch 'master' into lazy_tensor_staging

asuhan committed 3 years ago

771a5ae8

__torch_dispatch__: Populate kwargs dictionary with keyword-only arguments (#62822)

ezyang committed 3 years ago

e55f2718

Modify GHA CI to use PYTORCH_IGNORE_DISABLED_ISSUES based on PR body (#62851)

janeyx99 committed 3 years ago

2b83007a

[Static Runtime] Added a cache for NNC generated code across different calls to the same ops (#62921)

navahgar committed 3 years ago

8b54b14f

Enable upper for torch.linalg.cholesky (#62434)

Rong Rong (AI Infra) committed 3 years ago

3782f3ec

[nnc] Updated IR cloning to create clones of expressions in addition to statements (#62833)

navahgar committed 3 years ago

e54ee9ba

minor fixes in c10d for Windows (#62953)

peterjc123 committed 3 years ago

5deeaab3

Add handling of list write to remove mutation (#62904)

Elias Ellison committed 3 years ago

fff83f3f

Add tensor-scalar op (#62903)

Elias Ellison committed 3 years ago

254148ec

Port `sum.dim_IntList` kernel to structured kernels. (#61642)

ysiraichi committed 3 years ago

4c4c5b14

Adding collective quantization API (#62142)

marjanfariborz committed 3 years ago

c7db642a

Newer Older