P
Patrick Schratz
Researcher at University of Jena
Publications - 14
Citations - 429
Patrick Schratz is an academic researcher from University of Jena. The author has contributed to research in topics: Random forest & Hyperparameter optimization. The author has an hindex of 5, co-authored 13 publications receiving 179 citations. Previous affiliations of Patrick Schratz include Ludwig Maximilian University of Munich.
Papers
More filters
Journal ArticleDOI
Hyperparameter tuning and performance assessment of statistical and machine-learning algorithms using spatial data
TL;DR: In this article, the effects of spatial autocorrelation on hyperparameter tuning and performance estimation by comparing several widely used machine-learning algorithms such as boosted regression trees (BRT), k-nearest neighbor (KNN), random forest (RF) and support vector machine (SVM) with traditional parametric algorithms, such as logistic regression (GLM) and semi-parametric ones like generalized additive models (GAM) in terms of predictive performance.
Journal ArticleDOI
mlr3: A modern object-oriented machine learning framework in R
Michel Lang,Martin Binder,Jakob Richter,Patrick Schratz,Florian Pfisterer,Stefan Coors,Quay Au,Giuseppe Casalicchio,Lars Kotthoff,Bernd Bischl +9 more
TL;DR: The R (R Core Team, 2019) package mlr3 is a complete reimplementation of the mlr (Bischl et al., 2016) package that leverages many years of experience and learned best practices to provide a state-of-the-art system that is powerful, flexible, extensible, and maintainable.
Journal ArticleDOI
Predicting Forest Cover in Distinct Ecosystems: The Potential of Multi-Source Sentinel-1 and -2 Data Fusion
TL;DR: Sentinel-1 time series information improved forest cover predictions in open savanna-like environments with heterogeneous regional features and the presented approach proved to be robust and it displayed the benefit of fusing optical and SAR data at high spatial resolution.
Journal ArticleDOI
The performance of landslide susceptibility models critically depends on the quality of digital elevation models
TL;DR: Considering the critical importance of the quality of input data for landslide susceptibility, the authors investigated the performance improvements that can be achieved by different globally available digitized data sets, and concluded that the performance of different digitised data sets can be significantly improved.
Journal ArticleDOI
RQGIS: Integrating R with QGIS for Statistical Geocomputing
TL;DR: RQGIS supports the seamless integration of Python code using reticulate from within R for improved extendability, and offers a wider range of geoalgorithms, and is often easier to use due to various convenience functions.