caffe导入错误：即使在ubuntu 18.04上成功安装后，也是如此

Question

即使使用命令sudo apt install caffe-cpu成功安装了caffe，也出现导入错误。我可以在/usr/lib/python3/dist-packages/caffe处找到caffe文件（路径已添加到PYTHONPATH）。也安装了caffe目录的requirements.txt文件中提到的所有要求。

我正在使用Ubuntu 18.04 LTS，Python3。

有人可以帮助我解决此错误吗？

import caffe        

Traceback (most recent call last):
      File "6_reconstruct_alphabet_image.py", line 17, in <module>
        import caffe
      File "/usr/lib/python3/dist-packages/caffe/__init__.py", line 1, in <module>
        from .pycaffe import Net, SGDSolver, NesterovSolver, AdaGradSolver, RMSPropSolver, AdaDeltaSolver, AdamSolver, NCCL, Timer
      File "/usr/lib/python3/dist-packages/caffe/pycaffe.py", line 15, in <module>
        import caffe.io
      File "/usr/lib/python3/dist-packages/caffe/io.py", line 2, in <module>
        import skimage.io
      File "/usr/lib/python3/dist-packages/skimage/__init__.py", line 158, in <module>
        from .util.dtype import *
      File "/usr/lib/python3/dist-packages/skimage/util/__init__.py", line 7, in <module>
        from .arraycrop import crop
      File "/usr/lib/python3/dist-packages/skimage/util/arraycrop.py", line 8, in <module>
        from numpy.lib.arraypad import _validate_lengths
    ImportError: cannot import name '_validate_lengths'

Answer 1

问题已解决：由于caffe构建未成功完成，因此出现错误。我建议不要使用sudo apt install caffe-cpu命令（在Ubuntu的官方caffe安装指南中提到了该命令）。因为它将最终导致上述错误。最好从源代码安装。

让我提供逐步指导以在Ubuntu 18.04 LTS中成功安装caffe：

1] sudo apt-get install -y --no-install-recommends libboost-all-dev

2] sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libboost-all-dev libhdf5-serial-dev \ libgflags-dev libgoogle-glog-dev liblmdb-dev protobuf-compiler

3]git clone https://github.com/BVLC/caffecd caffecp Makefile.config.example Makefile.config

4]sudo pip install scikit-image protobufcd pythonfor req in $(cat requirements.txt); do sudo pip install $req; done

5]修改Makefile.config文件：取消注释行CPU_ONLY := 1和行OPENCV_VERSION := 3。

6]在Makefile中找到LIBRARIES行并将其更改为如下：

LIBRARIES += glog gflags protobuf boost_system boost_filesystem m hdf5_hl hdf5 \
        opencv_core opencv_highgui opencv_imgproc opencv_imgcodecs

7] make all

现在您可能会遇到类似这样的错误：

CXX src/caffe/net.cpp
src/caffe/net.cpp:8:18: fatal error: hdf5.h: No such file or directory
compilation terminated.
Makefile:575: recipe for target '.build_release/src/caffe/net.o' failed
make: *** [.build_release/src/caffe/net.o] Error 1

要解决此错误，请执行步骤8。

8]install libhdf5-dev打开Makefile.config，找到包含LIBRARY_DIRS的行并追加/usr/lib /x86_64-linux-gnu/hdf5/serial找到INCLUDE_DIRS并附加/usr/include/hdf5/serial/（根据此答案）重新运行make all

9] make test

10] make runtest

11] make pycaffe

现在您可能会遇到类似这样的错误：

CXX/LD -o python/caffe/_caffe.so python/caffe/_caffe.cpp
python/caffe/_caffe.cpp:10:31: fatal error: numpy/arrayobject.h: No such file or directory
compilation terminated.
Makefile:501: recipe for target 'python/caffe/_caffe.so' failed
make: *** [python/caffe/_caffe.so] Error 1

要解决此错误，请执行步骤12。

12]在Makefile.config中找到PYTHON_INCLUDE行，并进行如下更改：

`PYTHON_INCLUDE := /usr/include/python2.7 \
/usr/local/lib/python2.7/dist-packages/numpy/core/include`

13]通过将以下行添加到〜/ .bashrc文件的末尾，将模块目录添加到我们的$ PYTHONPATH中：

`sudo vim ~/.bashrc`
`export PYTHONPATH=$HOME/Downloads/caffe/python:$PYTHONPATH`
`source ~/.bashrc'

14]完成。

Answer 2

如果有任何较旧版本的软件包，可能会发生这种情况。特别是，根据Traceback，主要原因可能是scikit-image软件包的旧版本。只需使用命令升级软件包。

对于Python 2.7：pip install --upgrade scikit-image

和

对于Python 3.x：pip3 install --upgrade scikit-image