Anaconda 4.4.0 (Python 3.6 version) Write down the initial settings of Windows instead of a memo.
Windows 10 (64bit) Anaconda 4.4.0 (64bit)
Take the Anaconda pass. Added C: \ Users \ xxx \ Anaconda3 \ Scripts to the environment variable Path. (xxx sets your installation folder)
First, create a configuration file. It took me quite a while to get here.
> jupyter notebook --generate-config
.Jupyter / jupyter_notebook_config.py will be generated in the location where you installed Anaconda.
Change the folder where Jupyter starts first to any location you like.
c.NotebookApp.notebook_dir = 'D:\Sukina\basyo' #Describe the path of your favorite folder
c.NotebookApp.open_browser = False #Don't let the browser start up without permission
When I started Jupyter, the browser started up without permission and it was troublesome, so I set it to False. After launching Jupyter, access http: // localhost: 8888 and it's ok.
For those who launch Jupyter from shortcuts *** Important: Remove the% USERPROFILE% character at the end of the shortcut "Link to:". *** *** Otherwise, the config file doesn't seem to load. (I got stuck ...)
Anaconda 4.4.0 (2017/06/02 latest) There doesn't seem to be a Python 3.6 OpenCV package that can be installed with conda.
> conda install -c https://conda.binstar.org/menpo opencv3
Fetching package metadata .............
Solving package specifications: .
UnsatisfiableError: The following specifications were found to be in conflict:
- opencv3 -> python 2.7*
- python 3.6*
Use "conda info <package>" to see the dependencies for each package.
If you search on the net, you will find a combination of Python3.5 and OpenCV3.1. I'm going to build a 3.5 virtual environment and install OpenCV. I don't have any code that I want to run, so I'll try installing OpenCV3.
Build a Python 3.5 virtual environment with Anaconda. With create, you can easily install 3.5 and build a virtual environment. The environment name is py35. If you activate and change the environment, the prompt will change.
If you don't include anaconda in conda create, you just include Python 3.5 I changed it because it becomes.
~~> conda create --name=py35 python=3.5~~
> conda create --name=py35 python=3.5 anaconda
Fetching package metadata ...........
> activate py35
(py35)
(py35) conda install -c https://conda.binstar.org/menpo opencv3
When I installed it via conda again, I was able to install OpenCV safely. For the time being, confirm.
(py35) python
>>> import cv2
>>> cv2.__version__
'3.1.0'
You have successfully installed OpenCV 3.1.
[Addition] Add the kernel so that you can select it in Jupyter. Jupyter：[Kernel]->[Change Kernel]->[py35]
(py35) ipython kernel install --user --name=py35 --display-name=py35
For Jupyter kernel selection, I referred to Settings for easily selecting multiple kernels with Jupyter. Or rather, it is as it is.
Recommended Posts