Registry error regarding ConvAWS

[liaopeiyuan]

Thanks for your error report and we appreciate it a lot.

Checklist
--1. I have searched related issues but cannot get the expected help.--
--2. The bug has not been fixed in the latest version.--

Describe the bug
When reproducing x101 results, the following error is produced:
KeyError: 'ConvAWS is already registered in conv layer'

Reproduction

1. What command or script did you run?

 bash ./tools/dist_train.sh configs/reppoints_v2/reppoints_v2_x101_fpn_giou_mstrain_2x_coco.py 4

2. Did you make any modifications on the code or config? Did you understand what you have modified?
   I updated the path to coco dataset, and altered the num_classes in the bbox head (because my new dataset has fewer classes)

3. What dataset did you use?

Custom dataset that appears as a COCO dataset (works on the official mmdetection branch)

Environment

sys.platform: linux
Python: 3.7.6 | packaged by conda-forge | (default, Jun  1 2020, 18:57:50) [GCC 7.5.0]
CUDA available: True
CUDA_HOME: /tools/cluster-software/cuda-cudnn/cuda-10.2-7.6.5
NVCC: Cuda compilation tools, release 10.2, V10.2.89
GPU 0,1,2,3,4,5,6,7: Quadro RTX 8000
GCC: gcc (GCC) 5.2.0
PyTorch: 1.5.0
PyTorch compiling details: PyTorch built with:
  - GCC 7.3
  - C++ Version: 201402
  - Intel(R) Math Kernel Library Version 2019.0.4 Product Build 20190411 for Intel(R) 64 architecture applications
  - Intel(R) MKL-DNN v0.21.1 (Git Hash 7d2fd500bc78936d1d648ca713b901012f470dbc)
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - NNPACK is enabled
  - CPU capability usage: AVX2
  - CUDA Runtime 10.2
  - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_37,code=compute_37
  - CuDNN 7.6.5
  - Magma 2.5.2
  - Build settings: BLAS=MKL, BUILD_TYPE=Release, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -fopenmp -DNDEBUG -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DUSE_INTERNAL_THREADPOOL_IMPL -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, USE_CUDA=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_STATIC_DISPATCH=OFF,

TorchVision: 0.6.0
OpenCV: 4.2.0
MMCV: 1.0.2
MMDetection: 2.2.0+cb94bd1
MMDetection Compiler: GCC 5.2
MMDetection CUDA Compiler: 10.2

Error traceback
N/A, the rest is simply tracebacks of registry and imports

Bug fix
There seems to be something wrong with duplicates of ConvAWS in the registry.

[Scalsol]

Have you followed the INSTALL.md to install this repo? If so, the mmcv version should be 0.6.2, 1.0.2 is not compatible with this repo. You could run pip instal mmcv==0.6.2 to fix this problem.