AssertionError: Default process group is not initialized

[HaoweiGis]

Describe the bug
python tools/train.py configs/danet/danet_r50-d8_512x1024_40k_cityscapes.py. I get an error when using custom data for model training, AssertionError: Default process group is not initialized.
GPU now has two target detection networks running, is this the reason? mmdetection can train multiple networks simultaneously.

Environment info
sys.platform: linux
Python: 3.7.7 (default, Mar 23 2020, 22:36:06) [GCC 7.3.0]
CUDA available: True
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 10.1, V10.1.243
GPU 0: Tesla V100-PCIE-32GB
GCC: gcc (Ubuntu 9.3.0-10ubuntu2) 9.3.0
PyTorch: 1.5.0
PyTorch compiling details: PyTorch built with:

• GCC 7.3
• C++ Version: 201402
• Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 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 10.1
• 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 7.6.3
• 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: 0.6.0a0+82fd1c8
OpenCV: 4.2.0
MMCV: 1.0.2
MMSegmentation: 0.5.0+b72a6d0
MMCV Compiler: GCC 7.3
MMCV CUDA Compiler: 10.1

[HaoweiGis]

mmdetection and mmsegmentation can be installed in a docker container？

[xvjiarui]

Hi @HaoweiGis
If you would like to debug with non-distributed training, you need to change SyncBNto BN since distributed training is required by PyTorch SyncBN.

[HaoweiGis]

Hello@xvjiarui
Thank you for your reply, After change SyncBN to BN. The model is training.

[xvjiarui]

Hi @HaoweiGis
Thanks for letting us know. Please note that all models in this repo are trained with 4 GPUs.
If you are training with the same iterations on single GPU, the performance may drop.

[AbdullahRJafar]

I am having the same problem. How do I change SyncBN to BN?

[xvjiarui]

Hi @AbdullahRJafar
You may modify it in the config file.

[PriyankaJain-1998]

facing same issue, which file under configs folder?

[xiexinch]

facing same issue, which file under configs folder?

Hi @PriyankaJain-1998
At each config/_base_/models/xxx.py.
And you can also run tools/dist_train.sh by setting GPUS=1, like
./tools/dist_train.sh config.py 1