FCOS源码安装

官方Github上给出的安装步骤

Installation

Requirements:

• PyTorch >= 1.0. Installation instructions can be found in https://pytorch.org/get-started/locally/.
• torchvision
• cocoapi
• yacs
• matplotlib
• GCC >= 4.9,< 6.0
• (optional) OpenCV for the webcam demo

Option 1: Step-by-step installation

# first, make sure that your conda is setup properly with the right environment
# for that, check that `which conda`, `which pip` and `which python` points to the
# right path. From a clean conda env, this is what you need to do

conda create --name FCOS
conda activate FCOS

# this installs the right pip and dependencies for the fresh python
conda install ipython

# FCOS and coco api dependencies
pip install ninja yacs cython matplotlib tqdm

# follow PyTorch installation in https://pytorch.org/get-started/locally/
# we give the instructions for CUDA 10.2
conda install pytorch torchvision cudatoolkit=10.2 -c pytorch

export INSTALL_DIR=$PWD

# install pycocotools. Please make sure you have installed cython.
cd $INSTALL_DIR
git clone https://github.com/cocodataset/cocoapi.git
cd cocoapi/PythonAPI
python setup.py build_ext install

# install PyTorch Detection
cd $INSTALL_DIR
git clone https://github.com/tianzhi0549/FCOS.git
cd FCOS

# the following will install the lib with
# symbolic links, so that you can modify
# the files if you want and won't need to
# re-build it
python setup.py build develop --no-deps


unset INSTALL_DIR

# or if you are on macOS
# MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ python setup.py build develop

Option 2: Docker Image (Requires CUDA, Linux only)

The following steps are for original maskrcnn-benchmark. Please change the repository name if needed.

Build image with defaults (CUDA=9.0, CUDNN=7, FORCE_CUDA=1):

nvidia-docker build -t maskrcnn-benchmark docker/

Build image with other CUDA and CUDNN versions:

nvidia-docker build -t maskrcnn-benchmark --build-arg CUDA=9.2 --build-arg CUDNN=7 docker/

Build image with FORCE_CUDA disabled:

nvidia-docker build -t maskrcnn-benchmark --build-arg FORCE_CUDA=0 docker/

Build and run image with built-in jupyter notebook(note that the password is used to log in jupyter notebook):

nvidia-docker build -t maskrcnn-benchmark-jupyter docker/docker-jupyter/
nvidia-docker run -td -p 8888:8888 -e PASSWORD=<password> -v <host-dir>:<container-dir> maskrcnn-benchmark-jupyter

我自己的安装过程

# 创建conda虚拟空间FCOS
conda create -n FCOS python==3.7.5

# 激活FCOS环境
conda activate FCOS
# 安装ipython
conda install ipython

# 使用pip安装 ninja cython matplotlib tqdm
pip install ninja cython matplotlib tqdm

# 安装pytorch torchvision cudatoolkit
# 首先可以先看一下自己的cuda版本,我这里是cuda10.1 可能之后会升级为10.2
nvcc -V
# 安装对应版本的cudatoolkit
conda install pytorch torchvision cudatookit=10.1

不知道为什么老是会安装cpu版本的pytorch，之后需要改为gpu版本

# 下来需要安装cocoapi
# 首先从github上clone
export INSTALL_DIR=$PWD
cd $INSTALL_DIR
git clone https://github.com/cocodataset/cocoapi.git
# 下载完成后输入下面的命令
cd cocoapi/PythonAPI
python setup.py build_ext install

看到这个字样就是安装成功了

# 接下来clone FCOS代码
git clone https://github.com/tianzhi0549/FCOS.git
# 进入 FCOS文件
cd FCOS/
#运行下面这个命令
python setup.py build develop --no-deps

发现报错

解决办法

# 进入FCOS/fcos_core/csrc/cpu路径下
# 发现有三个文件，问题就出在nms_cpu.cpp和ROIAlign_cpu.cpp两个文件里
# 使用vim打开这两个文件
# 将头文件改为 vision.h
# 因为原代码引用的是“cpu/vision.h”在当前文件目录下，没有cpu这个文件，因此会报错找不到路径
# 可能这两个文件是在上层文件里吧，不知道出了什么错

# 重新运行上面的命令
python setup.py build develop --no-deps

安装成功