train_val: Split Habitat Data into training and Validation Data

In ailich/mytools: Stuff

Split Habitat Data into training and Validation Data

Description

Splits data into training and validation samples. Uses stratified random sampling without replacement. Will randomly draw the specified proportion without replacement from each class to generate validation data. Training set is the original data with the observations from validation data removed. Returns a list containing training data as the first element and validation data as the second element. The list elements will be the same data type as the input of my_data

Usage

train_val(my_data, hab_col, prop_val, seed = NULL)

Arguments

my_data	a spatialpoints dataframe containing, dataframe, or tibble containing all data
hab_col	number specifying the column to check
prop_val	proportion of samples to be used in validation
seed	sampling is initiated by set.seed(seed) if input is not NULL. This will generate consistent results

ailich/mytools documentation built on Jan. 7, 2023, 11:16 a.m.