Tuning Random Forest on Time Series Data

Manuel Tilgner Blog, Data Science

Training random forests on time series is one thing, but tuning them? It’s not like you can just apply cross validation and be done with it. Or can you? This post forms part two our mini-series “Time Series Forecasting with Random Forest”. Find out how you can tune the hyperparameters of the random forest algorithm when dealing with time series data. The answers might surprise you!


Dynamic UI Elements in Shiny

Oliver Guggenbühl Blog, Data Science

Shiny enables its users to quickly create a fixed UI with code. Although simple, this can prove to be quite limiting. Applying the principles of reactivity to the UI part of a ShinyApp is a natural progression from reactive coding as we know it already from Shiny’s server side. This blog entry reviews and discusses two of the most convenient tools for doing so.