Interpretable Machine Learning for Finance

Northwestern University

MLDS 490-0-1 Special Topics

Syllabus and Data

Syllabus and data sets will be available on the course Canvas.

Textbooks

Students are highly encouraged to purchase the following two books for the course:

• Molnar, Christoph. Interpretable Machine Learning: A Guide for Making Black Box Models Explainable. 3rd ed. https://christophm.github.io/interpretable-ml-book/. Accessed on March 28, 2025.

• Molnar, Christoph. Interpreting Machine Learning Models With SHAP: A Guide With Python Examples and Theory on Shapley Values. Independent publication, 2023.

Key Repo Files and Folders

• conda_env_requirements.yml : conda environment definition for the repository

• Interpretable ML.code-profile: configuration file for VS code profile

• IML4Finance.code-workspace: workspace file for the local repository

• course_utils/: shared Python helper functions used across course materials

• Data/: dataset-specific loading and preprocessing scripts for the course data sets

• Lectures/: Quarto source files (.qmd), rendered lecture and lab outputs (.html), and saved model/report artifacts used in class

System Prerequisites

• Miniforge3

• VS Code IDE

• Git

• GitHub Desktop

Clone the Repo

1. Open GitHub Desktop (and link your GitHub account).
2. Click on File > Clone Repository....
3. Select the URL tab.
4. Enter the URL of the repository
5. Click on Choose... and select the directory where you want to save the repository.
6. Click on Clone.

Conda Environment

1. In Windows, click the Start button.
2. Search for Miniforge Prompt.
3. Change the directory to the local repository (see Step 5 in the previous section).
4. Create the conda environment from the conda_env_requirements.yml file:

conda env create -f conda_env_requirements.yml

5. Activate the conda environment:

conda activate env_AutoGluon_202502

6. Launch VS Code:

code

VS Code Profile

1. Open VS Code
2. Type Ctrl + Shift + P
3. Type >Profiles: New Profile
4. Delete the profile that is called "Untitled"
5. Click on the drop-down arrow next to the "New Profile" button
6. Click Import Profile
7. Click Select File
8. Choose Interpretable ML.code-profile