strisunshinewentingwang/pytorch

Pull Requests Commits

Update if condition

peterjc123 committed 3 years ago

Verified 513c4330

Automatically set BUILD_SPLIT_CUDA for cpp exts

peterjc123 committed 3 years ago

Verified 42c6e98d

Fixed _out variants of linear algebra functions (#51560)

IvanYashchuk committed 3 years ago

7ca97768

[RPC] delete torch/csrc/utils/future.h (#51698)

rohan-varma committed 3 years ago

df3d1d93

[RPC] Refactor rref_context to not use utils::Future (#51697)

rohan-varma committed 3 years ago

3b118228

log newly added construction and runtime stats at randomly selected iterations (#51394)

zhaojuanmao committed 3 years ago

d0795ab3

add stats that can only be collected at runtime (#51386)

zhaojuanmao committed 3 years ago

c75fa39b

[DDP] Enhance warning for find_unused_params (#52385)

rohan-varma committed 3 years ago

0c46b6b3

[DDP] unittest for when params arent used in backward pass (#52384)

rohan-varma committed 3 years ago

c29e279f

[DataLoader] Change signature of Functional DataPipe (#52458)

ejguan committed 3 years ago

4ee5bc74

Enable min & max for Float16 & BFloat16 (#51244)

imaginary-person committed 3 years ago

3adc8f8c

[quant][graphmode][fx] Fix fp16 dynamic quant for functional linear (#52369)

jerryzh168 committed 3 years ago

fb9f8950

Add support for pow (#52374)

nikithamalgi committed 3 years ago

83feaebf

Add NNC support for aten::hardtanh (a hot operation in mobilenet v2/v3) (#52394)

huiguoo committed 3 years ago

d8b28579

Add onnxifi interface for set/get options (#52388)

Yinghai Lu committed 3 years ago

f4c33edb

[ROCm] missing template declarations for complex blas (#52472)

jeffdaily committed 3 years ago

82548f3a

Improvements for FX tracer (#52232)

jfix71 committed 3 years ago

65f6e665

[glow] Extending AOT config with two more fields (#5359)

khabinov committed 3 years ago

bb7e07ce

[ROCm] Enable test_ddp_hooks.py test cases (#52403)

KyleCZH committed 3 years ago

e0b6252d

Add arm64 binary build (#52443)

janeyx99 committed 3 years ago

89bc9a58

Revert D26299594: [PyTorch Mobile] 15KiB size reduction by reducing MaxTypeIndex from 256 to 32

malfet committed 3 years ago

22adea04

Revert nightly docker build cuda version to 11.1.1. (#52234)

xuzhao9 committed 3 years ago

2d435442

[iOS GPU] Fix max_pool_2d (#52431)

xta0 committed 3 years ago

49c90648

Make LLVM the default backend for TE (#52314)

navahgar committed 3 years ago

c7a70eec

enable mkldnn conv2d backward to support mkldnn tensor input (#48994)

XiaobingSuper committed 3 years ago

8f3ed60d

[PyTorch Mobile] 15KiB size reduction by reducing MaxTypeIndex from 256 to 32 (#51881)

dhruvbird committed 3 years ago

9e545329

Allow broadcasting against lerp weights. (#52319)

gchanan committed 3 years ago

983347fa

[BE] _get_torch_cuda_version should return tuple (#52409)

Rong Rong (AI Infra) committed 3 years ago

b52e2e60

Fix upsample bicubic2d batching handling on CPU. (#52389)

gchanan committed 3 years ago

f72b4b83

Enhance Tensor.unflatten to support -1 as the inferred size (#51955)

RockingJavaBean committed 3 years ago

c7b00058

Older