README.md

In Bradley-Butcher/BNLearnOR: Compute OR Plots for BNLearn Fit Objects

BNLearnOR

This library intends to simplify producing Odds Ratios from BNLearn fit objects relatively straightforward.

Usage

It requires a BNlearn fit object, and a configuration yaml file, and is used like so:

config <- yaml::read_yaml('config.yml')
Produce.OR(fit, config)

This will create one, or multiple, forest plots dependent on the configuration file.

Configuration File

The following is an example of the configuration file for the ALARM dataset:

yaml target_variable: 'BP' target_levels: ['LOW', 'HIGH'] target_reference: 'NORMAL' variable_levels: HIST: {'0':'FALSE', '1':'TRUE'} CVP: {'0':'LOW', '1':'HIGH', '2':'NORMAL', 'Reference':'NORMAL'}

The variables specified in variable levels will appear in the plot. Here two plots will be created, one: Normal vs. Low, and two: Normal vs. High.

Reference levels for each variable can be specified in the variable_levels, if none is provided, the first level specficied will be the reference.

Bradley-Butcher/BNLearnOR documentation built on July 4, 2019, 12:59 p.m.