bbl-package {bbl} | R Documentation |
Supervised learning using Boltzmann Bayes model inference, which extends naive Bayes model to include interactions. Enables classification of data into multiple response groups based on a large number of discrete predictors that can take factor values of heterogeneous levels. Either pseudo-likelihood and mean field inference can be used with L2 regularization, cross-validation, and prediction on new data. Woo et al. (2016) <doi:10.1186/s12864-016-2871-3>.
A typical workflow consists of the following steps:
Set up bbl
object.
Prepare a data frame containing data to be used as training set.
Create a main object using data as input argument
(bbl
).
Train the model.
See train
.
Perform cross-validation (crossval
)
to optimize regularization.
Make prediction on new data.
See predict
.
Maintainer: Jun Woo jwoo@umn.edu (0000-0003-3220-2064)
Other contributors:
Jinhua Wang [contributor]