Iteratively building a classifier requires a mix of skill, diagnostic ability and guesswork. I'll lay out a framework that helps you build reliable classifiers with greater confidence and less random guesswork. We'll review different ways to tackle a classification challenge, visual diagnostics that help identify sources of error and missing but exploitable information and tools to help you simplify your final solution. Tools demonstrated will include sklearn, YellowBrick, ELI5, pandas_profiling and skopt. The approaches we'll discuss will apply equally to regression challenges. Whilst this talk is aimed at Intermediate Data Scientists, people at the start of their career will benefit by having a clear process and by being introduced to new tools.

Slides linked here