环境
开发环境
- GPU服务器:pytorch1.13.0,python3.10,cuda==11.7
- nvidia-smi 查看 CUDA 版本
3. 官网查看对应cuda版本的PyTorch,使用生成命令行安装
3. 环境安装
conda create -n centerface python=3.7
conda activate centerface
pip3 install torch torchvision torchaudio -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install -U cython -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install -U 'setuptools>=18.0' -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install -U 'matplotlib>=2.1.0' -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install tqdm -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install -U progress -i https://pypi.tuna.tsinghua.edu.cn/simple
mkdir /data/coding/detect
COCOAPI=/data/coding/detect/cocoapi
git clone https://github.com/cocodataset/cocoapi.git $COCOAPI
cd $COCOAPI/PythonAPI
python setup.py install --user
make
CenterFace_ROOT=/data/coding/detect/centerface_pytorch
git clone http://developer.sourcefind.cn/codes/modelzoo/centerface_pytorch.git $CenterFace_ROOT
cd $CenterNet_ROOT/src/lib/external/
python setup.py install --user
make
pip install opencv-python
pip install numba -i https://pypi.tuna.tsinghua.edu.cn/simple
SDK 环境
- 参考 https://github.com/chenjun2hao/CenterFace.pytorch
- 开源训练数据:http://shuoyang1213.me/WIDERFACE/
数据
训练
测试