strisunshinewentingwang/pytorch

Pull Requests Commits

Codegen: TraceType only includes operators being registered (#68691)

peterbell10 committed 2 years ago

fa09099b

[quant] Add a e2e test for standalone module + custom backend_config_dict (#70152)

jerryzh168 committed 2 years ago

779f41a7

[quant][graphmode][fx] Add qat module mapping support in backend_config_dict (#70287)

jerryzh168 committed 2 years ago

ce86881a

[fx2trt] Add version check for ProfilingVerbosity bulider config (#70286)

jerryzh168 committed 2 years ago

65faf1a7

[PyTorch Edge][QNNPack] Tighten Step Height for Indirection Buffers (#70530)

salilsdesai committed 2 years ago

6bc06ec3

[nn] adaptive_avg_pool{1/2/3}d : Error on negative `output_size` (#70488)

kshitij12345 committed 2 years ago

7bfaa230

Remove backward ops for mkldnn convolution (#70467)

jbschlosser committed 2 years ago

e6c3aa38

[quant][fx][graphmode] Support standalone module in _convert_do_not_use (#70151)

jerryzh168 committed 2 years ago

cfc71f56

add BFloat16 support for AdaptiveAvgPool2d on CPU (#56902)

mingfeima committed 2 years ago

401a6b68

[Reinstate] Wishart distribution (#70377)

nonconvexopt committed 2 years ago

bc40fb56

make ProcessException pickleable (#70118)

epwalsh committed 2 years ago

14d3d29b

[PyTorch Edge][QNNPack] Enable Depthwise Specific Conv3d Kernel for Kernel Size 3x3x3 (#69315)

salilsdesai committed 2 years ago

9c742bea

[PyTorch Edge][QNNPack] Depthwise Conv3d mp8x27 (per-channel) Sse2 Kernel (#69314)

salilsdesai committed 2 years ago

3d4590d1

[PyTorch Edge][QNNPack] Depthwise Conv3d mp8x27 (per channel) Neon Kernel (#69313)

salilsdesai committed 2 years ago

821c085c

[PyTorch Edge][QNNPack] Depthwise Conv3d Weight Packing (#69312)

salilsdesai committed 2 years ago

15d44332

[PyTorch Edge][QNNPack] Depthwise Conv3d Indirection Buffer Setup (#69311)

salilsdesai committed 2 years ago

db37fd38

[PyTorch Edge][QNNPack] Refactor Computing Step Dimensions (#69310)

salilsdesai committed 2 years ago

9863cd57

[PyTorch Edge][QNNPack] Operator-Level Conv3d Tests (#69309)

salilsdesai committed 2 years ago

cea3eba6

[PyTorch] Add Enum to IValue Deepcopy (#69937)

salilsdesai committed 2 years ago

35251a55

softplus_backward: remove output arg (#70296)

kshitij12345 committed 2 years ago

36db5017

[Operator Versioning][Test] Use hypothesis for better test input data and broader coverage (#70263)

cccclai committed 2 years ago

18dd5cdb

[quant][fx][graphmode][be] Change the type for output of convert to be torch.nn.Module (#69959)

jerryzh168 committed 2 years ago

c6272116

Add testing across mem_formats to ModuleInfos (#69317)

mikaylagawarecki committed 2 years ago

fb78a319

Add Nondeterministic Tol to gradient test in test_modules (#69402)

mikaylagawarecki committed 2 years ago

14f4b91f

Added antialias flag to interpolate (CPU only, bicubic) (#68819)

vfdev-5 committed 2 years ago

d2abf3f9

fix typos in torch/csrc/deploy/README.md (#70494)

JimEverest committed 2 years ago

2b00dbbb

AdaptiveLogSoftmaxWithLoss no_batch_dim support (#69054)

george-qi committed 2 years ago

8af39b76

Fix docs rendering for nn.Module.named_modules() (#70491)

lithuak committed 2 years ago

0460324b

Remove backward op for slow dilated 3d convolution (#70068)

jbschlosser committed 2 years ago

fb736c77

[rnn,gru,lstm]cell : no batch dim (#70236)

kshitij12345 committed 2 years ago

2c67621a

Newer Older