First release

First release (1.0) of pybitup.

Bayesian inference toolbox

Bayesian inference using standard MCMC algorithms (RWMH, AMH, DRMH, DRAMH) but also more advanced techniques (Hamiltonian Monte Carlo, method based on an Ito stochastic differential equation (ISDE)). Data from several experimental sets, different models (within pybitup or external black-box) can be used with their systematic and uncertain parameters. Sampling from known distributions using one of those algorithms is also available.
Post process of the distribution are plotting of individual Markov chains, 1D ksdensity, 2D ksdensity and posterior propagation checks are saved and can be plotted.

Uncertainty Propagation

Uncertainty propagaiton is performed using Monte Carlo simulations directly from the Markov chains from the Bayesian Inference. Confidence interval can be evaluated.
Some features like propagating from known distribution, or building emulator is not yet available.