You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After a series of events in 2023, which included a training
workshop on [**Predictive Soil Spectroscopy**](https://soilspectroscopy.github.io/soilspec-workshop/) at the ACS International Meeting 2023 in St Louis - MO, USA, we made available a hands-on tutorial where we walk-through the reader on processing and predictive modeling steps that are currently employed in the OSSL. Common concepts and analysis are also introduced.
In this tutorial, you will find information on why soil spectroscopy should be a fit-for-purpose technology, why good predictions flow from good data, and what are the best practices for model building.
Further on, we walk the reader through examples of data import, preparation, visualization, spectra resampling, preprocessing, and compression.
Lastly, we conclude the guide with an introduction the the MLR3 modeling framework and basic Chemometrics, exploring aspects of data splitting and resampling, model optimization, benchmarking, and uncertainty estimation via conformal prediction.
We hope that this tutorial gives a good overview of predictive soil spectroscopy and helps people explore new ideas.