Principles of Data Science

A hands-on introduction to the principles and methods of data science. This course is designed to equip you with tools to begin extracting insights and making decisions from data in the real world, as well as to prepare you for further study in statistics, machine learning, and artificial intelligence. We will analyze and visualize data of different shapes and sizes (e.g., tabular, textual, hierarchical, geospatial). We will discuss common patterns and pitfalls of data analysis. We will build and evaluate machine learning models, focusing on general concepts (rather than specific methods), including supervised vs. unsupervised learning, training vs. testing error, hyperparameter tuning, and ensemble methods. The focus will be on intuition and implementation, rather than theory and math. Implementation will be in Python and Jupyter notebooks, using libraries such as pandas and scikit-learn. This course culminates in a project where you apply the ideas to a data science problem of your choosing. Website: http://dlsun.github.io/stats112 Prerequisite: CS 106a (or equivalent programming experience in Python). Note: All students must enroll in a discussion section that meets on Tuesdays and Thursdays in addition to the main lecture.