Skip to main content
Talk

Introduction to Python and Machine Learning at MIT IAP 2019

Dr. Omar Costilla Reyes taught a three-session, hands-on introduction to Python and machine learning at MIT in January 2019. The course ran January 14–16 during MIT’s Independent Activities Period, for beginners and researchers.

Anonymized word cloud of what participants expected to learn, shared on the eve of the January 2019 MIT IAP course. Credit: Image courtesy of Dr. Omar Costilla Reyes.

A hands-on course during IAP

MIT’s Independent Activities Period (IAP) is a January term of short courses and activities open across the institute. Dr. Costilla Reyes, then a postdoctoral researcher in the Department of Brain and Cognitive Sciences, offered “Introduction to Python and Machine Learning” as a three-session course on January 14–16, 2019, sponsored by the department. According to the MIT IAP listing, the sessions took place in room 46-3015, no prior knowledge was required, and participants brought laptops. In December he invited anyone interested in learning Python and machine learning to sign up, and on the eve of the opening session he shared an anonymized word cloud of what his participants hoped to learn.

What the course covered

The listing assigned one topic to each day: an introduction to machine learning, then unsupervised learning, then supervised learning. In supervised learning a model is trained on examples that come with known answers, such as images labeled by category, so that it can predict the answer for new cases; in unsupervised learning the data carry no labels, and the model looks for structure such as groups of similar items. Participants worked with Python and its machine-learning libraries by writing and running code rather than only listening, and in preparing the material he drew on a set of open teaching notes on Python and its scientific-computing libraries.

Participants’ goals and later teaching

In the word cloud, the largest words, including learn, machine, Python, use, basic, and data, suggest mostly practical, beginner-level goals: learning to use Python and machine-learning tools and understanding how they work. The course was one of several teaching contributions he made at MIT that year, including a computational tutorial on neural decoding of spike trains and local field potentials in April 2019, whose video and materials MIT later published through OpenCourseWare.

Topics