strisunshinewentingwang/pytorch

Pull Requests Commits

Format reducer.cpp, hpp (#58593)

rohan-varma committed 3 years ago

d83c5a5c

[fx][graph_drawer] Improve graph drawer coloring and tensor_meta handling (#58699)

jfix71 committed 3 years ago

6d97a80d

[codemod][dirsync] Apply clang-format

zertosh committed 3 years ago

5455df2b

Revert D28497967: [quant][fx][graphmode][refactor] Remove qconfig_map from Quantizer

Chillee committed 3 years ago

21a93340

[quant][fx][graphmode][refactor] Remove qconfig_map from Quantizer (#58455)

jerryzh168 committed 3 years ago

1cf8f7a4

[Reducer] Completely remove VariableIndex (#58592)

rohan-varma committed 3 years ago

62adf9e1

[Try] [PyTorch Edge] Trim unused code related to CUDA and HIP Interfaces (#58689)

dhruvbird committed 3 years ago

8e4fc006

Tag PyObject on TensorImpl per torchdeploy interpreter (#57985)

ezyang committed 3 years ago

773cfae9

Change native functions to take `c10::string_view` args instead of `std::string` (#57680)

kurtamohler committed 3 years ago

fe8e5eb2

[Caffe2] [Easy] Fix comment on caffe2_serialize_using_bytes_as_holder to reflect correct types

voznesenskym committed 3 years ago

d1d24304

[Pytorch Edge] NAME -> SCHEMA (#58604)

JacobSzwejbka committed 3 years ago

db67699a

enable torch.cpu.amp.autocast (#57386)

leslie-fang-intel committed 3 years ago

0ede83db

[quant][graphmode][fx] Move qat_swap_modules outside of Quantizer (#58454)

jerryzh168 committed 3 years ago

b6dcdeac

[PyTorch] Migrate TI usage in ATen/native/cpu to borrowing (#58303)

swolchok committed 3 years ago

fdc5dfdd

[PyTorch] Add TI::borrowing_nullary_op and use it (#58280)

swolchok committed 3 years ago

7c15d320

Enable the quantization on XPU devices (#54857)

chengjunlu committed 3 years ago

618be18a

Add `#pragma once` to CUDA foreach headers (#58209)

crcrpar committed 3 years ago

ce3788d6

[quant][fx][graphmode][refactor] Factor out generate_qconfig_map to qconfig_utils.py (#58453)

jerryzh168 committed 3 years ago

f879e70f

[static runtime] out variant for full_like (#58079)

ajyu committed 3 years ago

bf1c936e

Support aten::leaky_relu for TE (#58464)

Edvard Ghazaryan committed 3 years ago

5211eeb2

[quant][graphmode][fx] Quantize the output of statically quantized fp16 op in QuantizeHandler (#58445)

jerryzh168 committed 3 years ago

4668d09c

[Android]Removed dependency with AppCompat. (#58527)

H1Gdev committed 3 years ago

6edd49a8

[third-party] Update nccl to 2.9.8 (#58667)

malfet committed 3 years ago

d8412142

Add missing .to_sparse(ndim) gradient (#58413)

kylevedder committed 3 years ago

bbf92e61

Enable `ceil`, `floor`, `frac`, `round` & `trunc` for BFloat16 on CUDA (#57910)

imaginary-person committed 3 years ago

8a3d9962

[jit] Implement ScriptProfile to collect instruction profiles. (#57397)

zhxchen17 committed 3 years ago

034a238b

Adds grid_sampler to autocast fp32 list for 1.9 (#58679)

mcarilli committed 3 years ago

e8c6a650

Do not use TF32 matmul in linalg and DDP tests (#56114)

xwang233 committed 3 years ago

691c1391

Added statistic related to out variant nodes

Edvard Ghazaryan committed 3 years ago

a7f06e1e

turn off deadline for adagrad test

ngimel committed 3 years ago

056287ae

Newer Older