strisunshinewentingwang/pytorch

Pull Requests Commits

[ao][sparsity] Composability of fusion and sparsity (#74847)

HDCharles committed 2 years ago

8ac47291

Revert "Support masked sum on CSR tensors [CPU, CUDA]"

pytorchmergebot committed 2 years ago

31ed77b7

CUDA Kernels: Use per-operator headers (4/4) (#71215)

peterbell10 committed 2 years ago

35d4a805

CUDA Kernels: Use per-operator headers (3/4) (#71214)

peterbell10 committed 2 years ago

cd62bbf7

CUDA Kernels: Use per-operator headers (2/4) (#71213)

peterbell10 committed 2 years ago

9d4a76dc

Cuda 11.6 Disable failing tests (#75420)

atalman committed 2 years ago

622cff3e

[JIT] Move log_extract.py helper functions to torch.utils

davidberard98 committed 2 years ago

9d05ce60

[SR] Refactor memory planner to prepare for new algorithm (#74730)

Mike Iovine committed 2 years ago

b7682d35

[JIT] Add support for closed over inf

eellison committed 2 years ago

00d11de5

[easy][PTE] Ensure the stream points to the beginning before calling getFileFormat (#75437)

pavithranrao committed 2 years ago

8bf8b64b

[fx2trt] support for ne, logical_not, logical_and (#75444)

Wei Wei committed 2 years ago

0b1f8b03

[quant][fx] Move output share qparam with input ops to backend_config_dict (#75315)

jerryzh168 committed 2 years ago

7f0d7962

fix to map an undefined tensor back to a tensor list

davidberard98 committed 2 years ago

ad07b7c3

[SR] Do not manage tensors that escape scope via container (#74966)

Mike Iovine committed 2 years ago

2f98fa91

Adding integration of SSA into LazyTensor

Gamrix committed 2 years ago

f1db3e46

[PyTorchEdge] Backport from v9 flatbuffer to v8 pickle (#75201)

pavithranrao committed 2 years ago

3001bda3

[quant] adding const qualifier to Quantizer::equalTo (#75355)

terrychenism committed 2 years ago

d264824b

[ci] run upload-test-stats on self-hosted runner

suo committed 2 years ago

c1bd33cf

Remove flake8 pre-commit hook

kit1980 committed 2 years ago

e3f546af

update eigen submodule to latest release (3.4.0) with rocm fixes

acxz committed 2 years ago

f98881b1

[ONNX] Use fixed test input for flaky test

BowenBao committed 2 years ago

e60c403b

[SR] Update README (#75283)

Mike Iovine committed 2 years ago

c1de6b19

[FSDP][Easy] Fix return in docstrings

awgu committed 2 years ago

e48ecfda

Revert "[CUDA] Add fastAtomicAdd to scatter_add"

pytorchmergebot committed 2 years ago

339e45ba

[ci] add rery to getting job id

suo committed 2 years ago

8a6f0a25

Add new keys for Graphcore IPU (DispatchKey / Backend / DeviceType)

AnthonyBarbier committed 2 years ago

ce9e27a0

Extend CSR constructor to support batched indices and values

IvanYashchuk committed 2 years ago

c7ae23b5

Support masked sum on CSR tensors [CPU, CUDA]

pearu committed 2 years ago

5c28216a

Fix addmm_cpu for int64

lezcano committed 2 years ago

e0f9c69f

Update documentation for scatter_reduce

mikaylagawarecki committed 2 years ago

11f1fef9

Newer Older