use_linearBayes: get the prediction from a Bayesian linear regression

In iqspr: Inverse Molecular Design

Description

Get the prediction from a Bayesian linear regression. For a prediction other than from a Bayesian linear model, see use_Model.

Usage

use_linearBayes(newx = NULL, model = NULL)

Arguments

newx	is a matrix of features (e.g. fingerprints and/or physical descriptors), issued from a test/validation set, for which the properties (e.g. targeted chemical properties) are predicted.
model	is the set of parameters defining a trained Bayesian linear regression model, issued from get_linearBayes, to be used as a predictor.

Value

the predicted properties with associated variance.

Examples

## Not run: linearBayes_pred <- use_linearBayes(newx = newx,
model = c(list(w0),list(V0_inv),list(a0),list(b0)))
## End(Not run)

iqspr documentation built on Aug. 1, 2017, 9:02 a.m.