Introduction to Data Science with Python
- Free to audit
- Paid certificate
- Course certificate
- Intermediate
- 8 weeks
Introduction to Data Science with Python is HarvardX's course on using Python for data science and the first steps of machine learning. It works through regression models (linear, multilinear and polynomial) and classification models (k-nearest neighbours and logistic regression) using the pandas, NumPy, matplotlib and scikit-learn libraries, and covers the practical questions around them: choosing the right model complexity, preventing overfitting, regularization, assessing uncertainty, weighing trade-offs and evaluating models. The emphasis is on hands-on practice with real data science problems, using Python for modelling, statistics and storytelling, and the course is framed as preparation for further study of machine learning and artificial intelligence.
Harvard rates it intermediate. It is self-paced and runs eight weeks at 3 to 4 hours per week. The course is free to audit on edX, which gives access to the material for the duration of the course; a verified certificate costs $299.
What you’ll learn
- Build linear, multilinear and polynomial regression models in Python
- Fit classification models such as k-nearest neighbours and logistic regression
- Use pandas, NumPy, matplotlib and scikit-learn on real data
- Choose model complexity and apply regularization to avoid overfitting
- Evaluate models and weigh trade-offs and uncertainty
Who it’s for
People who already write some Python and want a structured first course in modelling and machine learning; not a first programming course.
Source: Harvard Online (opens in a new tab) · Verified · Report a change
More free Harvard courses
Every Harvard course we have verified, with cost and certificate terms compared.
Data ScienceAll free data science courses
Free and free-to-audit data science courses from every provider, ranked.