Course website: https://5harad.com/mse231/
Please follow the instructions below to ensure that your computing setup is sufficient for the course.
A Unix-like setup is required (e.g., Linux, OS X, or Cygwin). We primarily use R (RStudio is recommended) and Python 3.7 (direct download works; you can install Anaconda Python instead if you desire). We use the Tidyverse suite of packages in R for data manipulation and visualization. We also use Vowpal Wabbit (a fast online learning algorithm), and Amazon Elastic MapReduce (a web service for distributed computing).
JupyterLab, which runs in the browser, can be a convenient way to run Python (and R), and allows interleaving of explanatory text with executable code. We recommend you install JupyterLab with both Python 3 and R kernels.
Note: You must already have Python 3 installed in order to proceed. If you installed the Anaconda distribution in particular, you already have JupyterLab.
Follow the installation instructions (linked above as well) to install JupyterLab, using
pip as appropriate. (
conda will do if you have Anaconda Python installed.)
Now in your terminal run the following command by typing it and pressing return:
This will start running a web server that serves the JupyterLab application. This command should automatically open your default web browser to the JupyterLab page.
If you do not see JupyterLab open, read the output of the command you ran in the terminal. If Jupyter started correctly, it should have printed to the screen the URL you can use to access Jupyter.
By default, JupyterLab does not have the ability to execute R code. In order to use R with JupyterLab, you need to install the
R shell started from the terminal,
to install the package, followed by
to tell Jupyter where to find the kernel spec. (For more detail on these two steps, see here.)
Now, when you run
jupyter lab, you should see the option to create a notebook with the
The tidyverse is a collection of powerful libraries for working with data. These tools make common data manipulation and visualization tasks easy.
To install, open the interactive
R shell again and run:
The Python 3 kernel should be available by default when you open JupyterLab.
In rare cases, usually with Anaconda, you may have Python 3 installed but only be able to create notebooks with a Python 2 kernel. If this describes your situation, here is a potentially useful StackOverflow post about the issue.