Generate data sets to apply binary classifiers

Generate data sets to apply binary classifiers

Usage

sim_response_xy(
  n = 500,
  x_dist = purrr::partial(runif, min = -1, max = 1),
  y_dist = x_dist,
  relationship = function(x, y) x > y,
  noise = 0.2
)

Arguments

n

An intenger

x_dist

A random number generation function.

y_dist

A random number generation function.

relationship

A function specify the relationship between x, y and the response. A function f(x, y) need return a logical value.

noise

A number between 0 and 1.

Examples

set.seed(123)

df <- sim_response_xy(n = 500)

df
#> # A tibble: 500 × 3
#>    response       x        y
#>    <fct>      <dbl>    <dbl>
#>  1 FALSE    -0.425  -0.293  
#>  2 TRUE      0.577  -0.267  
#>  3 FALSE    -0.182  -0.426  
#>  4 TRUE      0.766  -0.840  
#>  5 TRUE      0.881  -0.269  
#>  6 FALSE    -0.909  -0.644  
#>  7 FALSE     0.0562  0.0721 
#>  8 TRUE      0.785   0.00790
#>  9 TRUE      0.103   0.890  
#> 10 TRUE     -0.0868 -0.317  
#> # … with 490 more rows

plot(df)