Trained Neural Networks on Simulated Data of Groundwater Heat Plume Characteristics

Dataset

DOI

Inference package for thermal plume prediction (v1.0.0). Contains pre-trained MLP and randomized neural network models, the ba-predict CLI, the CSV data used to train the models obtained via simulation by Fabian Böttcher, sample input files, and a Dockerfile. CPU-only (no GPU required). Install with: pip install ./code. Full source code (training scripts, data, tests) available at https://github.com/thomas-baratto/BA.

More information can be found in the README.md. Use persistent identifiers from Software Heritage (

) to cite individual files or even lines of the source code.

Identifier
DOI	https://doi.org/10.18419/DARUS-5815
Metadata Access	https://darus.uni-stuttgart.de/oai?verb=GetRecord&metadataPrefix=oai_datacite&identifier=doi:10.18419/DARUS-5815

Provenance
Creator	Baratto, Thomas
Publisher	DaRUS
Contributor	Baratto, Thomas; Pelzer, Julia; Schulte, Miriam; Böttcher, Fabian
Publication Year	2026
Rights	info:eu-repo/semantics/openAccess
OpenAccess	true
Contact	Baratto, Thomas (University of Stuttgart); Pelzer, Julia (University of Stuttgart); Schulte, Miriam (University of Stuttgart)

Representation
Resource Type	Dataset
Format	text/tab-separated-values; text/x-python; application/x-docker-file; application/zip; application/octet-stream; text/markdown
Size	8104866; 3633; 4012; 447168; 992; 5113; 2975; 9050; 4190; 4269; 3224; 329; 340; 72; 317031; 1251757; 2037; 15508; 14095; 2104; 739; 152739; 1207939; 561; 724; 6197; 482; 805; 8249; 2908; 3951; 6398; 1205; 6157; 4913; 9269; 6361; 12216; 1023
Version	3.0
Discipline	Chemistry; Computer Science; Computer Science, Electrical and System Engineering; Earth and Environmental Science; Engineering Sciences; Environmental Research; Geosciences; Natural Sciences