Forecasting the Future: An Introduction to Machine Learning for...
summary
Claim: you can train build and deploy a weather forecasting model entirely in native Ruby. Tools shown are iruby for Jupyter notebooks plus Numo, Daru, and Rumale gems. Data is NOAA Atlanta airport CSV ~20k rows since 1960s trimmed from ~48 columns to five predictors. Model is multilinear regression with an 80/20 train/test split predicting next‑day max temperature. Result is a working demo with predictions and a GitHub project; persistence/export of the model is noted but not fully implemented.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:48Agenda and overview
- 2:29ML tools in Ruby
- 4:16Key Ruby ML libraries
- 5:00Weather data collection
- 5:44Data cleaning
- 9:0080-20 rule in data science
- 9:41Splitting data train/test
- 10:47Linear regression
- 13:17Training the model
- 14:22ML in Rails apps
- 16:11Resources and project
- 17:37Q&A
related talks
01h 16mPredicting Weather Patterns using Time Series ML
29mConnecting the Dots: Unleash the magic of AI in IOT
39mForecasting With Classical and Machine Learning Methods
01h 26mApplying Statistical Modeling & Machine Learning to Perform Time-Series Forecasting
42mRobust Automated Forecasting in Python and R
32mMachine Learning using Open Source R
22mPredict the future (or don't) with machine learning in applications
37mForecasting using machine learning methods