installing_scikit.md

Installing Scikit

It's better to use a virtual environment to install these packages to avoid conflicts with already installed versions. First install pip and virtualenv. In ubuntu use $ sudo apt-get install python-pip python-dev build-essential $ sudo pip install --upgrade pip $ sudo pip install --upgrade virtualenv

1. cd to the project
2. Create a virtual environment by virtualenv -p python3 virtual
3. cd virtual/bin and source activate to start the virtual environment
4. install numpy using pip pip install numpy
5. Install scipy using pip pip install scipy
6. Install SciKit using pip pip install -U scikit-learn
7. cd ../../ to come back to root project directory
8. Try something eg: start python by python and from sklearn import datasets ('' If success your'e all set")

Read here [http://www.pythonforbeginners.com/basics/how-to-use-python-virtualenv] for more about virtualenv.

OR with the virtualenvprop.txt file I have created you can easily set up all.

1. Clone project / Pull
2. Install virtualenv same as above.
3. virtualenv -p python3 virtual
4. source virtual/bin/activate (Now you'll see (virtual)$ in left of the terminal)
5. pip install -r virtualenvprop.txt
6. Try something eg: start python by python and from sklearn import datasets ('' If success your'e all set")

Viola you are now ready. Read the official documentation[http://scikit-learn.org/stable/documentation.html]

Please don't forget to add the create virtualenv folder to .gitignore if you used some else name for it.