George Brocklehurst, Development Director of thoughtbot New York, presents an Intro to Machine Learning workshop. In the first half, he answers what is Machine Learning and what can it do, discusses supervised vs unsupervised machine learning problems, and some of the tools and techniques used.
George then goes on to do a live-coding demonstration of a classification model
in which he builds a data set, engineers some features, and finally trains a
neural network; all to over-engineer the most complex way to solve fizzbuzz!
Supporting materials, such as the code used to generate the examples in the slides and a copy of the fizzbuzz implementation, can be found in this repo.
Check out more workshops, meetup presentations, editor tips/shortcuts, lightning talks, AMA’s, behind the scenes bloopers, and more by subscribing to the thoughtbot YouTube Page; and don’t hesitate to reach out and learn how thoughtbot can help you with your next Machine Learning project!