Create a Training Data Object

Generate a training data set from a original parent data set typically via some subset of the original parent. Final groups must be binary, and generate a 2 factor level "response" column used in many downstream statistical testing functions.

is.tr_data() checks whether an object is a tr_data class object.

Usage

create_train(data, ..., group_var, classes = NULL)

is.tr_data(data)

# S3 method for class 'tr_data'
plot(x, ft, main = ft, do_pdfs = FALSE, do_log = TRUE, ...)

Arguments

data: A data.frame used to create a training data set.
...: Arguments passed to dplyr::filter() used to subset rows. If passing to the S3 plot method for tr_data class objects, additional arguments are passed to SomaPlotr::plotCDFbyGroup() via the ....
group_var: character(1). Can be quoted or unquoted. Must be a column name of data.
classes: Either NULL, where no factor conversion will be performed (default), or a string character(2) indicating first and second class labels respectively. See the Details section for more information about factor levels.
x: A tr_data object.
ft: character(1). The name of a column in data containing values to generate CDFs or PDFs.
main: character(1). Title for the plot. See ggplot2::ggtitle().
do_pdfs: logical(1). Should smoothed densities PDF be plotted?
do_log: logical(1). Should values be log10-transformed?

Value

A tibble with an additional tr_data class. This object contains the subset training data with a additional attributes about the groupings and the "response" variable.

Logical. Whether data inherits from class tr_data.

Details

When specifying filtering variables, the factor levels will be ordered alphabetically in the resulting "response" variable unless ordering is specified by the classes argument. This is important, for example, when performing repeated univariate statistics where \(class2 - class1\), i.e. the positive class is 2!

Functions

plot(tr_data): Plots a CDF, and optionally an accompanying smoothed PDF for a specific feature in a "tr_data" object.

Author

Stu Field

Examples

# New "tr_data" object with default factor levels
classes <- c("setosa", "versicolor")
tr <- create_train(iris, Species %in% classes, group_var = Species)
tr
#> ══ Training Data Object ═══════════════════════════════════════════════
#> • response        Species
#> • class labels    'setosa', 'versicolor'
#> • counts          [50, 50]
#> • factor          TRUE
#> • n               2
#> ───────────────────────────────────────────────────────────────────────
#> # A tibble: 100 × 5
#>    Sepal.Length Sepal.Width Petal.Length Petal.Width Species
#>  *        <dbl>       <dbl>        <dbl>       <dbl> <fct>  
#>  1          5.1         3.5          1.4         0.2 setosa 
#>  2          4.9         3            1.4         0.2 setosa 
#>  3          4.7         3.2          1.3         0.2 setosa 
#>  4          4.6         3.1          1.5         0.2 setosa 
#>  5          5           3.6          1.4         0.2 setosa 
#>  6          5.4         3.9          1.7         0.4 setosa 
#>  7          4.6         3.4          1.4         0.3 setosa 
#>  8          5           3.4          1.5         0.2 setosa 
#>  9          4.4         2.9          1.4         0.2 setosa 
#> 10          4.9         3.1          1.5         0.1 setosa 
#> # ℹ 90 more rows

# Getting Variables
attr(tr, "response_var")
#> [1] "Species"

attr(tr, "class_labels")
#> [1] "setosa"     "versicolor"

attr(tr, "counts")
#>     setosa versicolor 
#>         50         50 

# with re-naming factors
tr2 <- create_train(iris, Species %in% classes,
                    group_var = Species, classes = rev(classes))
#> • Note: Class order is non-alphabetic: 'versicolor > setosa'
tr2
#> ══ Training Data Object ═══════════════════════════════════════════════
#> • response        Species
#> • class labels    'versicolor', 'setosa'
#> • counts          [50, 50]
#> • factor          TRUE
#> • n               2
#> ───────────────────────────────────────────────────────────────────────
#> # A tibble: 100 × 5
#>    Sepal.Length Sepal.Width Petal.Length Petal.Width Species   
#>  *        <dbl>       <dbl>        <dbl>       <dbl> <fct>     
#>  1          5.1         3.5          1.4         0.2 versicolor
#>  2          4.9         3            1.4         0.2 versicolor
#>  3          4.7         3.2          1.3         0.2 versicolor
#>  4          4.6         3.1          1.5         0.2 versicolor
#>  5          5           3.6          1.4         0.2 versicolor
#>  6          5.4         3.9          1.7         0.4 versicolor
#>  7          4.6         3.4          1.4         0.3 versicolor
#>  8          5           3.4          1.5         0.2 versicolor
#>  9          4.4         2.9          1.4         0.2 versicolor
#> 10          4.9         3.1          1.5         0.1 versicolor
#> # ℹ 90 more rows
# S3 plot method
ft <- "Sepal.Length"  # random feature
plot(tr, ft)
#> Registered S3 method overwritten by 'SomaPlotr':
#>   method   from      
#>   plot.Map SomaDataIO


plot(tr, ft, cols = c("blue", "red"))


plot(tr, ft, cols = c("black", "black"))   # b/w


plot(tr, ft, do_pdfs = TRUE)


plot(tr, ft, do_pdfs = TRUE, cols = c("blue", "red"))