strisunshinewentingwang/pytorch

Pull Requests Commits

Add Enum TorchScript serialization and deserialization support (#42963)

gmagogsfm committed 4 years ago

f269fb83

Workaround bugs in user side embedding meta info and better msgs (#43355)

Yinghai Lu committed 4 years ago

aa53b2d4

[quant][graphmode][fx] Add support for layer_norm (#43376)

jerryzh168 committed 4 years ago

aec917a4

[quant][graphmode][fx] Add support for elu (#43375)

jerryzh168 committed 4 years ago

089bb1a8

[quant][graphmode][fx] Add support for hardswish (#43374)

jerryzh168 committed 4 years ago

5a02c6b1

Mobile backward compatibility (#42413)

iseeyuan committed 4 years ago

93f1b5c8

[quant][graphmode][fx] Add support for mul and mul relu (#43373)

jerryzh168 committed 4 years ago

e96871ea

Revert D23252335: Refactor Vulkan context into its own files. Use RAII.

Mike Ruberry committed 4 years ago

6c772515

Fix enum constant printing and add FileCheck to all Enum tests (#42874)

gmagogsfm committed 4 years ago

8eb3de76

[quant][grapphmode][fx][test][refactor] Refactor quantized add test (#43372)

jerryzh168 committed 4 years ago

ff454cc4

[quant][graphmode][fx] Add support for batchnorm relu (#43335)

jerryzh168 committed 4 years ago

109ea59a

[quant][graphmode][fx] Add support for batchnorm (#43334)

jerryzh168 committed 4 years ago

9e87a8dd

Refactor Vulkan context into its own files. Use RAII. (#42273)

Ashkan Aliabadi committed 4 years ago

054073c6

[quant][graphmode][fx] Add support for cat (#43333)

jerryzh168 committed 4 years ago

3d76f706

[quant][graphmode][fx] Add support for add relu (#43332)

jerryzh168 committed 4 years ago

26be4dcf

[quant][graphmode][fx] Add support for add (#43331)

jerryzh168 committed 4 years ago

452a4737

.circleci: Prefer using env-file for docker run (#43293)

seemethere committed 4 years ago

6e48c88e

Normalize loops with non-zero start. (#43179)

navahgar committed 4 years ago

100649d6

Revert D23242101: [pytorch][PR] Implement first draft of autograd benchmark.

albanD committed 4 years ago

74781ab5

[quant][graphmode][fx] Add support for conv module + relu (#43287)

jerryzh168 committed 4 years ago

650590da

[quant] Enable from_float for quantized Embedding_Bag (#43176)

supriyar committed 4 years ago

3293fdfa

[quant] Make offsets an optional argument (#43090)

supriyar committed 4 years ago

b354b422

[quant] Create nn.quantized.dynamic.EmbeddingBag (#43088)

supriyar committed 4 years ago

4db8ca11

[TensorExpr] Simplify conditional select (#43350)

asuhan committed 4 years ago

f20a04fa

Fix PackedGemmMatrixFP16 repacking (#43320)

Yinghai Lu committed 4 years ago

743cff4a

Adds arccos, arcsin, arctan aliases (#43319)

Mike Ruberry committed 4 years ago

e57b89c8

Enables bfloat16 x [float16, complex64, complex128] type promotion (#43324)

Mike Ruberry committed 4 years ago

3aec1185

[jit] PyTorchStreamReader::getAllRecord should omit archive name prefix (#43317)

Dmytro Dzhulgakov committed 4 years ago

478fb925

Add Enum convert back to Python object support (#43121)

gmagogsfm committed 4 years ago

0bd35de3

Do not define the macro "isnan" (#43242)

xuhdev committed 4 years ago

f4b6ef9c

Newer Older