Signal Processing and Speech Communication Laboratory

Python and SciPy stack installation

Windows installation

Installing python and libraries that will be used in our course is made easy with Anaconda. It includes Python, IPython and the packages Numpy, Scipy, Matplotlib, Scikit-learn and others

  1. Download Anaconda graphical installer for Python 3.5 from https://www.continuum.io/downloads
  2. Run the graphical installer
  3. It is advised to tick both boxes: add Anaconda to PATH environment and register as default python

Linux installation

Using Anaconda

  1. Download Anaconda linux installer for Python 3.5 from https://www.continuum.io/downloads
  2. Run bash Anaconda3-2.5.0-Linux-x86_64.sh
  3. When it asks:

     Do you wish the installer to prepend the Anaconda3 install location
     to PATH in your /home/anand/.bashrc ? [yes|no]
     [no] >>> yes
    

    Answer “Yes”

Using your distribution package manager

Mac OS X installation

Install using Anaconda Graphical Installer for Python 3.5

Installing using Macports may take a very long time (depending on your system configuration), since it compiles a lot of the packages from scratch.

Installing an IDE

You are free to use your favorite IDE. Just make to configure Python 3.5 as the default Python interpreter for your course related projects in case you have conflicting python versions.

If you do not know what IDE to choose, we recommend to use PyCharm. It is dedicated to Python, includes an IPython command line, and many useful features to debug your code. A free version is available and more than sufficient for this course.

Testing installation

If you encounter any failure while testing your installation you probably have an older version of Python or IPython that is in conflict with the new one:

  1. Open the recently installed IPython. When opening the window, the version of Python used is specified in the header. Make sure you are using Python 3.5
  2. In the command line, try to import the libraries we are using in the course by typing: “import numpy,scipy,sklearn,matplotlib”.

Check your version (Should be 3.4 or 3.5)

$ python --version
Python 3.5.1

Support

Send a message to the newsgroup for help with installation tu-graz.lv.ci or visit one of us during office hours.