strisunshinewentingwang/pytorch

Pull Requests Commits

[caffe2] Enable fp16 for SparseNormalize op (#45551)

jianyuh committed 3 years ago

5c67cc7a

Add some more docs to expecttest. (#46263)

ezyang committed 3 years ago

2118d58d

[pytorch][glow][NNPI] Using int32 as indices for embedding_bag operators (#45878)

Allan Di Wu committed 3 years ago

1c3e335c

Avoid computing AutogradKey if not needed. (#46252)

Ailing Zhang committed 3 years ago

a37f2749

Complex autograd doc fix (#46258)

anjali411 committed 3 years ago

ac245f6b

[quant][fx][graphmode] Add prepare_custom_config_dict and convert_custom_config_dict (#46223)

jerryzh168 committed 3 years ago

67a0c0af

Automated submodule update: FBGEMM (#46271)

facebook-github-bot committed 3 years ago

dac68072

[quant][graph][fix] Set type for GetAttr nodes in remapTypes (#46250)

supriyar committed 3 years ago

95ccf34f

Use storage.cpu() for moving storage to CPU in serialization. (#46028)

t-vi committed 3 years ago

7b7f2519

.circleci: Fix android publish snapshot job (#46266)

seemethere committed 3 years ago

fc846db6

[quant][refactor] Alphabetize the entries in the quantized import (#46218)

z-a-f committed 3 years ago

5604997b

Support pytest for distribution testing (#45648)

neerajprad committed 3 years ago

faa9c22a

trying to make pow work for tensor raised to the power of a scalar (#46185)

janeyx99 committed 3 years ago

ad376f1a

Add torch._foreach_maximum(TensorList, TensorList) & torch._foreach_minimum(TensorList, TensorList) APIs (#45692)

Iurii Zdebskyi committed 3 years ago

1a57b390

Define the record_stream method in native_functions.yaml (#44301)

chengjunlu committed 3 years ago

5741de88

Refactor dispatcher and native to use Signature structure. (#45990)

ezyang committed 3 years ago

d705083c

Remove unnecessary byte-for-byte compatibility code that is not needed. (#45975)

ezyang committed 3 years ago

f0860326

Rename legacy_dispatcher to native. (#45974)

ezyang committed 3 years ago

8d5c899b

Reorder dispatcher/legacy_dispatcher types (#45973)

ezyang committed 3 years ago

527a8bee

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

Newer Older