BaM v1.1.1

• Fixed a bug affecting calibration with 0 data
• Modified GitHub action building Windows executable

Full Changelog: v1.1.0...v1.1.1

BaM v1.1.0

• GitHub actions to automatically compile BaM exe for linux and windows (failing on MacOS at the moment)
• New versions of the MAGE Model (hydraulics): MAGE_TEMP (to be deprecated) and MAGE_ZQV
• New command line arguments: --dontrun, --onerun and --saveprior
• Write a file summarizing the inference setup
• Various minor fixes and improvements

Full Changelog: v1.0.0...v1.1.0

BaM v1.0.0

First release of BaM (Bayesian Modeling), a framework aimed at estimating a model using a Bayesian approach and using it for prediction, with a particular focus on uncertainty quantification.

BaM takes the form of an executable file, piloted by a few configuration text files. It is a computational engine and mostly aims at generating MCMC samples from the posterior distribution associated with the model and the data, and to use these MCMC samples to make predictions.

The best place to start is probably the R package RBaM, which provides the friendliest user interface to pilot BaM. Lower-level information, in particular regarding the content and format of configuration files, can be found here. A report is also available in French.