There are a number of improvements required to the plotting routines. Some of this can probably be achieved through arviz already, for example to make useful corner plots of the parameters grouped appropriately. Posterior-predictive plots in more than 2 dimensions and for uncertain data are also essential.
There are a number of improvements required to the plotting routines. Some of this can probably be achieved through arviz already, for example to make useful corner plots of the parameters grouped appropriately. Posterior-predictive plots in more than 2 dimensions and for uncertain data are also essential.