strisunshinewentingwang/pytorch

Pull Requests Commits

Add NativeFunctionGroup (#45918)

ezyang committed 3 years ago

944eb0e3

Rewrite implementation of faithful cpp signatures (#45890)

ezyang committed 3 years ago

9079aea1

fix #45552 - adding add_done_callback(fn) to torch.futures.Future (#45675)

bdhirsh committed 3 years ago

a3caa719

Workaround for bug in DistributedDataParallel (#46186)

Flamefire committed 3 years ago

282f4ab9

[iOS][GPU] Add Metal/MPSCNN support on iOS (#46112)

xta0 committed 3 years ago

a277c097

[quant][fx][graphmode][api] Change API for custom module (#45920)

jerryzh168 committed 3 years ago

7f6a1b2b

Revert D24165889: Update native_functions.yaml to add DefaultBackend.

Ailing Zhang committed 3 years ago

e6d30c89

Update native_functions.yaml to add DefaultBackend. (#45938)

Ailing Zhang committed 3 years ago

1f9ddf64

Use const-references in nodes_to_rewrite range loop

malfet committed 3 years ago

ba1e0a88

Fix JIT test config (#46230)

malfet committed 3 years ago

4ad47156

Fix incorrect CUDA `torch.nn.Embedding` result when max_norm is not None and indices are not sorted (#45248)

kurtamohler committed 3 years ago

66505b64

[fx] use a linked list for nodes (#45708)

zdevito committed 3 years ago

88dcb95e

Adding information how to control randomness with DataLoader (#45749)

VitalyFedyunin committed 3 years ago

31ee5d8d

[pytorch][PR][Gradient Compression] Reduce the peak memory of fp16 compression provided by ddp comm hook (#46078)

Yi Wang committed 3 years ago

ee3d3e6d

[pt][quant] Support either min or max in qclamp (#45937)

dskhudia committed 3 years ago

87a4baf6

Implement ravel (#46098)

ejguan committed 3 years ago

bed3b405

fix test_serialization not working with Windows. (#46120)

Rong Rong committed 3 years ago

b98e3594

[NNC] Fix two more bugs in Cuda Half support (#46129)

nickgg committed 3 years ago

f3db6877

adding complex support for distributed functions and . fix #45760 (#45879)

bdhirsh committed 3 years ago

c02efdef

clean up dataclasses installation to only <3.7 (#46182)

Rong Rong committed 3 years ago

8de9aa19

including tensorexpr tests in CI for all configs (#46188)

janeyx99 committed 3 years ago

ba78eb80

[caffe2] add PlanExecutorTest ErrorPlanWithCancellableStuckNet (#46110)

dahsh committed 3 years ago

85c3ba55

Made exception message for torch.LongTensor() legacy constructor more readable (#46147)

partypyro committed 3 years ago

8d5256e6

Improve error checking of Storage._writeFile. (#46036)

gchanan committed 3 years ago

2070834b

Fix error in Binomial to retain lazy logit initialization (#46055)

neerajprad committed 3 years ago

9202c443

Fix typing errors in the torch.distributions module (#45689)

xuzhao9 committed 3 years ago

146721f1

[quant][test] Add mul_scalar test (#46106)

jerryzh168 committed 3 years ago

6a001dec

Only use hacky_wrapper_for_legacy_signatures if an op needs it (#45742)

smessmer committed 3 years ago

6ba6ecb0

Fix mkldnn build on legacy x64 arch (#46082)

pinzhenx committed 3 years ago

e1f74b18

[fix] torch.kthvalue : handle non-contiguous CUDA tensor (#45802)

kshitij12345 committed 3 years ago

a8142316

Newer Older