ImportError: numpy.core.multiarray failed to import

[onurays]

I am trying to setup this project for hours on Mac OS. Could anyone please help me to resolve this import error in server.py.

ImportError: numpy.core.multiarray failed to import Traceback (most recent call last): File "./server.py", line 28, in import cv2 ImportError: numpy.core.multiarray failed to import

[kaikuehne]

Hi, did you install numpy? You also need to install scipy, pandas, scikit-learn and scikit-image and the packages listed in demos/web/requirements.txt.

Note: If you are using a virtualenv (which I recommend), the cv2 package may not be available because the opencv python module was only installed for the system-python. In that case, just copy the cv2 module into your virtualenv:

cp /usr/local/lib/python2.7/site-packages/cv* <VIRTUALENV>/lib/python2.7/site-packages

[onurays]

Yes, I installed all of the dependencies. cv2 is also available. I already copied them to virtualenv.

numpy-1.10.1 pandas-0.17.0 scipy-0.16.0 scikit_image-0.11.3 scikit_learn-0.16.1

[bamos]

Do the solutions here help? http://stackoverflow.com/questions/28157976/importing-opencv-and-getting-numpy-core-multiarray-failed-to-import

[bamos]

See also http://stackoverflow.com/questions/20518632/importerror-numpy-core-multiarray-failed-to-import

[onurays]

I already tried both of your suggestions. numpy and others are up to date. Thanks.

[bamos]

What numpy version are you using? I'm using 1.9.2.

~$ python2 -c 'import numpy; print (numpy.version.version)'
1.9.2

---

Edit - sorry, I see you're using 1.10.1 from above.

[bamos]

Just as a sanity check, can you try outputting the numpy version right before the failed cv2 import?

[bamos]

https://github.com/Itseez/opencv/issues/4893 shows importing numpy before cv2 helps in some cases

[onurays]

Actually, i just deleted everything to start over, this will take some moment. Thanks a lot. I hope, import order is the case. Will write here, after test.

[onurays]

After struggle of another couple hours, I have a clean setup by following these tutorial for Mac OS: http://www.pyimagesearch.com/2015/06/15/install-opencv-3-0-and-python-2-7-on-osx/

Now, I am stuck on dlib setup. It is not compiling: CMake Error at /usr/local/Cellar/cmake/3.3.2/share/cmake/Modules/FindBoost.cmake:1245 (message): Unable to find the requested Boost libraries.

Unable to find the Boost header files. Please set BOOST_ROOT to the root directory containing Boost or BOOST_INCLUDEDIR to the directory containing Boost's headers. Call Stack (most recent call first): /Users/onurays/Documents/openface/dlib-18.16/dlib/add_python_module:57 (FIND_PACKAGE) CMakeLists.txt:5 (include)

[bamos]

• [x] Sorry, the README doesn't mention dlib's boost dependency. I'll add instructions within the next few days.

[bamos]

Sorry for the bad setup experience. Depending on your use case, the Docker container works well on OSX with an Ubuntu Docker machine.

-Brandon.

On Tue, Oct 20, 2015 at 4:44 AM onurays notifications@github.com wrote:

For MacOS users: "brew install boost-python".

And now server.py throws "Segmentation fault: 11". Seems like an OpenCV bug for MacOS. I think i will give up :)

— Reply to this email directly or view it on GitHub https://github.com/cmusatyalab/openface/issues/30#issuecomment-149479507 .

[RapidR3D]

I just wanted to say that I have been having this issue for the past two days and began after an installation of visual studio 2015. I'm running windows 10, python 3.5. After much trial and error the solution was as simple as going to the sitepackages directory and manually deleting anything that had to do with NumPy or MatPlotLib. After deleting I reinstalled using the command: python -m pip install --user matplotlib-2.0.0-cp35-cp35m-win_amd64.whl

Everything is now back up and running. Including Jupyter Notebook. Which was an odyssey in itself as well. I hope this can help someone else. It seems so simple now but I swear I nearly through my rig out into the yard over trying to figure out a viable fix for this problem.

[bamos]

Thanks for the helpful post @RapidR3D!