strisunshinewentingwang/pytorch

Pull Requests Commits

Rewrite ShardedTensor.gather to use dist.gather instead of gather_object (#77272)

Rodrigo Kumpera committed 2 years ago

668599a6

Revert "Python function to extract information on mobile::Module from flatbuffer (#77328)"

pytorchmergebot committed 2 years ago

5e3e5a54

[Re-submit] Make tracer be able to trace different forward functions

sijiac committed 2 years ago

efcbbb17

Revert "[JIT] during freezing, cast optional bias to half if weight is half"

pytorchmergebot committed 2 years ago

246078e2

Revert "Add Sparse Compressed tensor support to torch.clone"

pytorchmergebot committed 2 years ago

f1c8e8fa

[BE] Move numba pinned version to requirements-ci.txt

malfet committed 2 years ago

4596ecb4

Re-land BUCK build for pytorch mobile (#77612)

linbinyu committed 2 years ago

1f804956

Fix doc about type promotion of lshift and rshift (#77613)

zasdfgbnm committed 2 years ago

09751746

Change test_sparse_csr test signatures (#77595)

cpuhrsch committed 2 years ago

89e32f52

Move other div variants to upgraders map

tugsbayasgalan committed 2 years ago

31d9f7c3

Add Sparse Compressed tensor support to torch.clone

pearu committed 2 years ago

20ba6e69

[JIT] during freezing, cast optional bias to half if weight is half

davidberard98 committed 2 years ago

2547be51

Support comparing Sparse Compressed tensors

pearu committed 2 years ago

9a608af8

[GH1] Allow conclusions to include workflows and checks (#77597)

janeyx99 committed 2 years ago

f9aaf9d3

[shard] add clone/detach and set requires_grad for ShardedTensor

wanchaol committed 2 years ago

25fa964d

Revert "Revert "[LT] Codegen ReuseNode for supported ops""

desertfire committed 2 years ago

25c6ebd1

check that flip doesn't accept repeating dimensions (#77500)

ngimel committed 2 years ago

375e21b2

Use FastAtomicAdd for index_add/_index_reduce

mikaylagawarecki committed 2 years ago

fe1232a4

[mergebot cli] add new revert regex (#77459)

zengk95 committed 2 years ago

906fc777

Add Sparse CSC support to torch.empty

pearu committed 2 years ago

d76efed5

Replace SparseCsrTensorImpl crow/col_indices with compressed/plain_indices

pearu committed 2 years ago

77f1ca00

stirde_properties fix (#77460)

jjsjann123 committed 2 years ago

1b1dec45

Bugfix gradient formula for index_reduce('prod') + separate out sample_inputs for index_reduce

mikaylagawarecki committed 2 years ago

7ba4e124

Compressed sparse layout conversion stubs (#77489)

cpuhrsch committed 2 years ago

8c608a79

Revert "enable NVFuser by default"

pytorchmergebot committed 2 years ago

2a905aef

skip primTorch nvfuser tests on rocm (#77468)

jjsjann123 committed 2 years ago

8875453d

[Model Averaging] Support disabling post-local gradient sync (#76723)

wayi1 committed 2 years ago

5ab8afe4

[coreml] Use special throw macro when encountering CoreML API errors (#77429)

SS-JIA committed 2 years ago

93d84c0f

[metal] Use special throw macro when encountering Metal API errors (#77430)

SS-JIA committed 2 years ago

c34be4eb

[GHA] Use silent checkout (#77547)

malfet committed 2 years ago

46e16737

Newer Older