Doing Data Science with Python 2

It starts where real projects start but tutorials rarely do — setting up a working environment with Anaconda, Jupyter and Git — then moves through extracting data from databases, APIs and web scraping, and on to exploring and processing it. Seeing the whole workflow join up is the point: it's the connective tissue between skills people usually learn in isolation, and the part that turns 'I know pandas' into 'I can run a project'.

It's less useful if you can't yet code Python, or if you're after deep ML modelling specifically, since it's about the lifecycle rather than the algorithms. It's on a Pluralsight subscription (Standard about $29/month or $299/year, Premium about $45/month or $499/year, 10-day trial). The workflow it teaches stays stable even as libraries change (as of 2026).

This course walks through a complete data science project lifecycle: setting up a working environment (Anaconda, Jupyter, Git), extracting data from databases/APIs/web scraping, exploring and processing data (statistics, missing values, outliers, feature engineering), and building, evaluating, and deploying predictive models — closing with model persistence and exposing the trained model as a Flask API endpoint.