viviBartMatrix function

viviBartMatrix

Returns a matrix or list of matrices. If type = 'standard' a matrix filled with vivi values is returned. If type = 'vsup' two matrices are returned. One with the actual values and another matrix of uncertainty values. If type = 'quantiles', three matrices are returned. One for the 25

viviBartMatrix(
  trees,
  type = "standard",
  metric = "propMean",
  metricError = "CV",
  reorder = FALSE
)

Arguments

• trees: A data frame created by extractTreeData function.
• type: Which type of matrix to return. Either 'standard', 'vsup', 'quantiles'
• metric: Which metric to use to fill the actual values matrix. Either 'propMean' or 'count'.
• metricError: Which metric to use to fill the uncertainty matrix. Either 'SD', 'CV' or 'SE'.
• reorder: LOGICAL. If TRUE then the matrix is reordered so high values are pushed to the top left.

Returns

A heatmap plot showing variable importance on the diagonal and variable interaction on the off-diagonal.

Examples

if(requireNamespace("dbarts", quietly = TRUE)){
 # Load the dbarts package to access the bart function
 library(dbarts)
 # Get Data
 df <- na.omit(airquality)
 # Create Simple dbarts Model For Regression:
 set.seed(1701)
 dbartModel <- bart(df[2:6], df[, 1], ntree = 5, keeptrees = TRUE, nskip = 10, ndpost = 10)

 # Tree Data
 trees_data <- extractTreeData(model = dbartModel, data = df)

 # VSUP Matrix
 vsupMat <- viviBartMatrix(trees = trees_data,
                           type = 'vsup',
                           metric = 'propMean',
                            metricError = 'CV')
 }