Machine Learning with F#, Redux
summary
Argues that .NET + F# is a practical ML stack and shows it end to end: uses IPython notebooks with an F# kernel and F# type providers to load Titanic CSVs, Plotly for charts, and trains models with Accord.NET logistic regression and k‑means clustering. Demonstrates feature engineering via a typed feature DSL, parallelizable map/filter pipelines, and deployment paths to Azure. Validates with cross‑validation shows overfitting (train error ~2% vs validation ~40%) so feature choice and metrics drive model quality.
AI-written summary · about
chapters (8)
- 0:00Start
- 5:35A well-documented dataset
- 25:34Highlights
- 27:21Choose your own adventure
- 28:00What is your question?
- 29:05In machine learning terms
- 42:55Learning from data
- 58:41Dirty secret of machine learning
related talks
59mMachine Learning in .NET (ML.NET)
01h 03mMachine learning for C# developers: Introducing ML
27mN Ways to Do Machine Learning on .NET
50mA Pragmatic Deep Dive into ML.NET
52mML.NET - Machine Learning made for .NET
51mAgile experiments in Machine Learning with F#
57mMachine Learning Made Easy – An Introduction to ML.NET
27mUsing C# for Data Science and Machine Learning