A introduction to causal inference using common tools from the python data stack
You'll need graphviz for our first exercise notebook, to visualize causal graphs.
- Linux:
- Depending on your distro, here are the possible commands
- MacOS:
- You can easily install it via homebrew:
brew install graphviz
- You can easily install it via homebrew:
- Windows:
- The graphviz.org website has convenient installers for you
In your terminal, use git to clone the repo to your machine.
git clone git@github.com:ronikobrosly/pydata_nyc_2022.gitIf you are less comfortable with git, there is an easy alternative: You can simply download a zip file of it here :)
Now you'll need to ensure you have a working python environment set up.
This tutorial requires python version 3.9 for the notebook exercises to properly work.
Create a new virtual environment for this tutorial. You can do this a number of ways. If you are running Anaconda python, you can do this through the conda command. Another method is to use the virtualenv python package. If you are a Windows or Linux user, please use this guide. If you are a MacOS user you can follow this guide.
Name your environment causal_modeling
"Activate" this environment (see the above guide) and then run the following command in the root folder of this repo:
pip install -r requirement.txt
This will install all the necessary packages for the tutorial.
As an optional step, you can try to run the check_environment.py file (in the root folder of the repo)
while within your virtual environment. You can do so by running python check_environment.py in your terminal. It will alert you if you're missing any required python packages.
Once the above is complete, you'll need to run the following commands:
python -m ipykernel install --user --name causal_modeling --display-name "Python (causal_modeling)"
In the terminal, execute jupyter lab.
Navigate to the notebooks directory and open your notebook of choice. You will probably be asked which environment you would like to use with the notebook. Select the Python (causal_modeling) environment you created in the step above.
Slides to go along with the tutorial are available as a Google Doc Presentation.
I love would to hear your feedback on these tutorial materials! Please send your comments to roni.kobrosly@gmail.com.