Sailing with Scicloj: A Bayesian Adventure
summary
Predicts boat speed from wind angle and wind speed by combining synthetic theory and summer measurements, using Clojure data tools to build empirical polar diagrams and a Bayesian model via Scicloj + cmdstan-clj. Data collection runs on Raspberry Pis and plots use Hanami/Plotly; the model returns 80% credible surfaces over the angle/wind grid. A sample measurement at 17.5° and 10.6 kn gave 4.1 kn versus a synthetic peak near 5.2 kn, revealing actionable gaps between theory and real performance.
AI-written summary · about
chapters (11)
- 0:00Introduction and context
- 1:42The sailing data challenge
- 3:21Understanding polar charts
- 5:05Working with synthetic data
- 8:03Regression modeling basics
- 11:44Refining the polynomial model
- 16:11Bayesian statistics approach
- 19:57Probabilistic programming
- 22:26Analyzing model results
- 25:07Quantified sailing insights
- 27:51Clojure ecosystem and wrap-up
related talks
31mBayesian Data Analysis of Nonparametric Models in Clojure
26mDoing data science with Clojure: the ugly, the sad, the joyful
38mScientific Clojure, a bird's eye view
47mQuick and Dirty Data Science with Incanter
34m"Real @toms with Clojure!"
40m"Clojure in the Fintech Ecosystem"
24mFunctional Physics Code Along with Visual Blocks
41mPredicting Vehicle Usage with Clojure