preprocessPredictionData: Preprocess covariates. DataFrames will be preprocessed based...

In stochtree: Stochastic Tree Ensembles (XBART and BART) for Supervised Learning and Causal Inference

Preprocess covariates. DataFrames will be preprocessed based on their column types. Matrices will be passed through assuming all columns are numeric.

Description

Preprocess covariates. DataFrames will be preprocessed based on their column types. Matrices will be passed through assuming all columns are numeric.

Usage

preprocessPredictionData(input_data, metadata)

Arguments

input_data	Covariates, provided as either a dataframe or a matrix
metadata	List containing information on variables, including train set categories for categorical variables

Value

Preprocessed data with categorical variables appropriately handled

Examples

cov_df <- data.frame(x1 = 1:5, x2 = 5:1, x3 = 6:10)
metadata <- list(num_ordered_cat_vars = 0, num_unordered_cat_vars = 0, 
                 num_numeric_vars = 3, numeric_vars = c("x1", "x2", "x3"))
X_preprocessed <- preprocessPredictionData(cov_df, metadata)

stochtree documentation built on April 4, 2025, 2:11 a.m.