The models from CRAIMODELS are trained using CRAI (https://github.com/FREVA-CLINT/climatereconstructionAI) with different climate datasets.
Information about the training datasets and training parameters are given in craimodels/models/info-models.yml
.
The trained models can be used for evaluation with:
- CRAI command line interface (CLI):
crai-evaluate
- CRAI Python library:
from climatereconstructionai import evaluate evaluate()
For more information about the installation and usage of CRAI, see https://github.com/FREVA-CLINT/climatereconstructionAI/blob/main/README.md.
CRAIMODELS
is licensed under the terms of the BSD 3-Clause license.
CRAIMODELS
is maintained by the Climate Informatics and Technology group at DKRZ (Deutsches Klimarechenzentrum).
- Current contributing authors: Étienne Plésiat.