Fast, Easy, and Visual Bayesian Inference
Group together latent parameters by prior distribution.
causact: Fast, Easy, and Visual Bayesian Inference
Check if 'r-causact' Conda environment exists
Create a graph object for drawing a DAG.
Convert graph to Diagrammer object for visualization
Add dimension information to causact_graph
Add edge (or edges) between nodes
Generate a representative sample of the posterior distribution
Merge two non-intersecting causact_graph objects
Add a node to an existing causact_graph object
Generate a representative sample of the posterior distribution
Create a plate representation for repeated nodes.
Render the graph as an htmlwidget
Plot posterior distribution from dataframe of posterior draws.
probability distributions
Install causact's python dependencies like numpyro, arviz, and xarray.
Store meaningful parameter labels
The magrittr pipe
The Bernoulli Distribution
Set DiagrammeR defaults for graphical models
Accelerate Bayesian analytics workflows in 'R' through interactive modelling, visualization, and inference. Define probabilistic graphical models using directed acyclic graphs (DAGs) as a unifying language for business stakeholders, statisticians, and programmers. This package relies on interfacing with the 'numpyro' python package.
Useful links