maskrcnn-benchmark入坑指南
https://github.com/facebookresearch/maskrcnn-benchmark
1.環(huán)境安裝
指定python3.7的版本
conda create --name maskrcnn_benchmark python==3.7
conda remove -n maskrcnn_benchmark --all
https://github.com/facebookresearch/maskrcnn-benchmark/blob/master/INSTALL.md
2.
python3.7
cuda9.0
export PATH=$PATH:/usr/local/cuda-9.0/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-10.0/lib64
conda create --name maskrcnn_benchmark -y
conda activate maskrcnn_benchmark
# this installs the right pip and dependencies for the fresh python
conda install ipython pip
# maskrcnn_benchmark and coco api dependencies
pip install ninja yacs cython matplotlib tqdm opencv-python
# follow PyTorch installation in https://pytorch.org/get-started/locally/
# we give the instructions for CUDA 9.0
conda install -c pytorch pytorch-nightly torchvision cudatoolkit=9.0
export INSTALL_DIR=$PWD
# install pycocotools
cd $INSTALL_DIR
git clone https://github.com/cocodataset/cocoapi.git
cd cocoapi/PythonAPI
python setup.py build_ext install
# install cityscapesScripts
cd $INSTALL_DIR
git clone https://github.com/mcordts/cityscapesScripts.git
cd cityscapesScripts/
python setup.py build_ext install
# install apex
cd $INSTALL_DIR
git clone https://github.com/NVIDIA/apex.git
cd apex
python setup.py install --cuda_ext --cpp_ext
# install PyTorch Detection
cd $INSTALL_DIR
git clone https://github.com/facebookresearch/maskrcnn-benchmark.git
cd maskrcnn-benchmark
python setup.py build develop
unset INSTALL_DIR
3.
編譯問(wèn)題
_C.cpython-36m-x86_64-linux-gnu.so: undefined symbol: _ZN3c106Device8validateEv
原因的是maskrcnn-benchmark沒(méi)有編譯好
nvcc查看自己的路徑
which nvcc
/usr/bin/nvcc 這不是cuda9.0的nvcc所以報(bào)錯(cuò)
解決辦法
export PATH=/usr/local/cuda-9.0/bin:$PATH
cd /home/v-zhiwwa/HOI/maskrcnn/maskrcnn-benchmark && python setup.py build develop
4. 檢測(cè)一張圖片
export PATH=/usr/local/cuda-9.0/bin:$PATH
export PYTHON=$PYTHON:/home/v-zhiwwa/HOI/maskrcnn/maskrcnn-benchmark/build/lib.linux-x86_64-3.7
cd /home/v-zhiwwa/HOI/maskrcnn/maskrcnn-benchmark/demo && python wzw.py
5.檢測(cè)結(jié)果
圖片的目標(biāo)有bbox和分割的邊界
