1 安装caffe
https://blog.csdn.net/yhaolpz/article/details/71375762
这个博主写的已经足够详细,一般按照该步骤安装的话一般不会出现安装不成功的情况
2 按照上篇博客安装在最后一步在命令行import caffe时会出现一些疑问,比如在python命令时导入成功,但是在python3命令下不能成功,这种情况很常见,一般没有python2 和python3都安装caffe的必要,但是个人就是比较好奇
由于caffe在一个文件下只能安装一个python,所以我在另一个目录下又新建了一个caffe文件,比如在我在~/caffe中是python2的路径,在~/Documents/caffe下是python3的路径
对Makefile.config中除了如安装caffe博客一样的修改之外,有如下修改
# We need to be able to find Python.h and numpy/arrayobject.h.
#PYTHON_LIBRARIES := boost_python3 python3.5m
#PYTHON_INCLUDE := /usr/include/python2.7 \
#/usr/lib/python2.7/dist-packages/numpy/core/include
# Anaconda Python distribution is quite popular. Include path:
# Verify anaconda location, sometimes it's in root.
# ANACONDA_HOME := $(HOME)/anaconda
# PYTHON_INCLUDE := $(ANACONDA_HOME)/include \
# $(ANACONDA_HOME)/include/python2.7 \
# $(ANACONDA_HOME)/lib/python2.7/site-packages/numpy/core/include
# Uncomment to use Python 3 (default is Python 2)
PYTHON_LIBRARIES := boost_python3 python3.5m
PYTHON_INCLUDE := /usr/include/python3.5m \
/usr/lib/python3.5/dist-packages/numpy/core/include之后make all 即可
上面的做法只是为了安装python3的caffe,但是虽然python2和python3的caffe都已安装成功,但是并不能共存,由于路径的原因,
所以在使用时,在~/.bashrc中
export PYTHONPATH=~/caffe/python:$PYTHONPATH,则python在命令行可以import caffe成功
使用
export PYTHONPATH=~/Documents/caffe/python:$PYTHONPATH,则python3在命令行可以import caffe成功
3 pycharm import caffe失败问题
永久性解决
https://blog.csdn.net/LOOKTHEWIND/article/details/71499098
当然也会涉及到python2和python3的问题
4 Check failed: status == CUDNN_STATUS_SUCCESS (4 vs. 0)
一般是权限问题,很有可能是在安装caffe是使用的命令是sudo make all
但也有可能是其他问题,可以修改caffe的访问权限,更好的方式是在执行.py文件时,加上sudo命令即可解决问题
5 使用sudo .py 或者sudo ./pycharm.sh运行pycahem但是 import caffe失败
https://blog.csdn.net/jane_6091/article/details/84570236
还是python2 和python3对应
持续更新...