Introduction to Machine Learning

Introduction to Machine Learning#

Overview#

Welcome to Intro to Machine Learning, a course designed to introduce you to core machine learning concepts. You will explore the machine learning landscape, and the main stages of the machine learning pipeline. We will cover model selection, error, evaluation and validation, and put all of these into practice using Python. We will introduce Scikit-learn, and use it to pre-process data, train models, and build pipelines.

Course objectives#

  • Explain the main theoretical concepts of machine learning (at a high level), the machine learning landscape, and be able to provide some examples of machine learning applications.

  • Train basic linear models using scikit-learn, building understanding of model-based learning.

  • Understand concepts such as model selection, error, evaluation, validation, and be able to put these into practice with Python.

  • Explain the main stages of the machine learning pipeline, and be able to create a pipeline running on real-world data.

Prerequisites#

  • Both Introduction to Python and Python for Data Analysis are pre-requisites for attending this course. If you have not attended these courses, please review the full course materials in your own time.

  • As a minimum, you should be comfortable performing data analysis in Numpy and Pandas, and in creating plots with Matplotlib, without the need for regular guidance or support. Familiarity with Jupyter Notebooks is highly recommended.

  • In addition, we also recommend that you are comfortable using virtual environments: please see our self-study guide to do this.

Session content#

This course will be delivered via a mix of slides and programming tutorials. Materials can be found at the following links:

Session 1#

Session 2#

Session 3#

Contents