Examples

The following notebooks demonstrate the main workflows and features of rxmc. They are rendered with their pre-computed outputs; to run them locally, install the example dependencies first:

pip install -ve '.[examples]'
jupyter lab

Basic calibration

• Calibration of a line
• Clearly to make a model, we need to understand these things called Observations.
• Nice!
• Introduction to the likelihood
• Comparing likelihood models: there is a right way, and many wrong ways!
• Multiple constraints
• Multiple constraints with offset domain
• Robust likelihoods: Student-t vs the multivariate normal

Realistic nuclear physics calibration

• 30s optical potential calibration
• From a measurement to a calibrated potential
• Calibrating an optical potential with CalibrationConfig: emcee vs dynesty

Advanced topics

• Correlated observations
• Shared systematics between cross-section datasets
• Model discrepancy with a Gaussian-process term
• GP discrepancy on a differential cross section
• Bayesian calibration of polynomials, including inference of overall normalization
• Covariance-structure gallery
• Comparison of sampling algorithms for calibration of a line with unknown model error
• Large N implies overconfidence? Do we need to re-scale the Likelihood?