Explore projects

Massively parallel reactive transport simulator exploring acceleration strategies such as embedding of AI/ML and cache of results in Distributed Hash Tables. Developed in cooperation with computer scientists of University of Potsdam.

An R-package for accessing GTS2 data using the R programming language

This is the core of the auto ML approach that does model fitting, validation, and selection mostly using caret interfaces. It was written for automatically finding suitable machine learning models for table regression tasks.

http://swc-bb.gitext.gfz-potsdam.de/swc-lessons/2020-04-20-potsdam-berlin/Python/

R code used in the paper "DecTree v1.0 - Chemistry speedup in reactive transport simulations: purely data-driven and physics-based surrogates" submitted to Geoscientific Model Development (gmd2020-445)

https://swc-bb.gitext-pages.gfz-potsdam.de/swc-pages/2020-12-08-virtual/

http://swc-bb.gitext.gfz-potsdam.de/swc-lessons/2019-11-26-potsdam-berlin/shell/

http://swc-bb.gitext.gfz-potsdam.de/swc-lessons/2019-11-26-potsdam-berlin/git/

2 half-day | Python | Beginner | Online | Link

https://swc-bb.git-pages.gfz-potsdam.de/swc-lessons/2021-04-27-GFZ-virtual/Python

GCex is an interactive exploration prototype for geochemical simulation data. It implements an early version of Stacked Parameter Relation (SPR) visualization which guides the user to interesting relationships in the data from the model.