strisunshinewentingwang/pytorch

Pull Requests Commits

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

peterbell10 committed 2 years ago

f04396ac

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

jerryzh168 committed 2 years ago

649f9b17

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

jerryzh168 committed 2 years ago

c836dab9

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

jerryzh168 committed 2 years ago

1b318a20

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

salilsdesai committed 2 years ago

e559d3ba

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

kshitij12345 committed 2 years ago

5546f937

Remove backward ops for mkldnn convolution (#70467)

jbschlosser committed 2 years ago

f10e2760

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

jerryzh168 committed 2 years ago

bb4d7eb5

add BFloat16 support for AdaptiveAvgPool2d on CPU (#56902)

mingfeima committed 2 years ago

3bc38971

[Reinstate] Wishart distribution (#70377)

nonconvexopt committed 2 years ago

7033b67b

make ProcessException pickleable (#70118)

epwalsh committed 2 years ago

02d50973

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

salilsdesai committed 2 years ago

2f9d53eb

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

salilsdesai committed 2 years ago

e86293dc

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

salilsdesai committed 2 years ago

8ada477b

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

salilsdesai committed 2 years ago

959acbd9

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

salilsdesai committed 2 years ago

c4bfbfef

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

salilsdesai committed 2 years ago

281e061d

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

salilsdesai committed 2 years ago

9c388943

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

salilsdesai committed 2 years ago

44d0f3a5

softplus_backward: remove output arg (#70296)

kshitij12345 committed 2 years ago

bd2652c1

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

cccclai committed 2 years ago

7f726d59

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

jerryzh168 committed 2 years ago

297176f4

Add testing across mem_formats to ModuleInfos (#69317)

mikaylagawarecki committed 2 years ago

474e0a94

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

mikaylagawarecki committed 2 years ago

84ef3a7f

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

vfdev-5 committed 2 years ago

7ad91c15

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

JimEverest committed 2 years ago

2cae226c

AdaptiveLogSoftmaxWithLoss no_batch_dim support (#69054)

george-qi committed 2 years ago

9916a6ed

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

lithuak committed 2 years ago

29573caf

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

jbschlosser committed 2 years ago

c146f7c6

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

kshitij12345 committed 2 years ago

d0b83e1e

Newer Older