weight_sparsity_histogram

weight_sparsity_histogram#

mofaflex.pl.weight_sparsity_histogram(model, bins=50, nrow=None, ncol=None)#

Plot a histogram of probabilities that weights are non-zero for views with SnS prior.

The spike-and-slab prior is a mixture distribution of a Normal distribution with a potentially non-zero mean and a Dirac delta distribution at zero. This function plots the posterior probability of the Normal mixture compoonent.

Parameters:

  • model (MOFAFLEX) – The MOFA-FLEX model.

  • bins (int (default: 50)) – Number of histogram bins.

  • nrow (int | None (default: None)) – Number of rows in the faceted plot. If None, plotnine will determine automatically.

  • ncol (int | None (default: None)) – Number of columns in the faceted plot. If None, plotnine will determine automatically.