Go to Top
Investigating the polar regions from the inside out

Homework Exercise #1: Installation of Python packages

Follow the below instructions to complete the installation process.  To test the installation, students should run a set of test scripts which are available from the file conda_installation_test.zip that can be downloaded HERE

For questions or issues with installation, please contact instructor Fabian Walter (walter@vaw.baug.ethz.ch)

UPDATE for Mac users who have recently updated their OS: A bug has been reported where updating the Mac OS results in a failure of jupyter's automatic opening.  A student has provided information on a fix.  For more information click HERE



Download Miniconda (Python 3.6 distribution):




Follow installation instructions:




Windows: open a Command Prompt window.

Mac/Linux: Open a terminal. If not already active, enter bash shell. Stay

in bash shell from now on.


> bash


Make sure that conda is in the PATH variable.


Mac/Linux: $ which conda

Windows: > where conda


Install needed packages and put everything into environment called "colorado".


$ conda config --add channels conda-forge

$ conda config --add channels obspy

$ conda create --name colorado python=3 obspy numpy scipy pandas jupyter


Activate colorado environment. Note the "(colorado)" prefix in the

bash shell prompt from now on.


$ source activate colorado


Create and enter working directory.


$ mkdir workdir

$ cd workdir


Unpack conda_installation_test.zip.


$ unzip conda_installation_test.zip

(in Windows right-click on .zip, then "Extract All")


Run installation tests.

Mac/Linux:$ ./test_installation.sh

Windows: > python test_installation.py


Note that the following message can be ignored:


Gtk-Message: Failed to load module "canberra-gtk-module"


Running the installation test will tell you if all needed packages were

correctly installed. As a final check, a jupyter notebook should open up in

your brower. On Windows, you have to do this final step manually via the

following command:


> jupyter notebook