f_predict_plot_regression_distribution: Plot distribution of model predictions vs observed

In erblast/oetteR: Collection of personal R functions

Description

takes a dataframe with predictions and a title column and returns a list with one violin and one histogram plot to compare distributions.

Usage

f_predict_plot_regression_distribution(data, col_title = "title",
  col_pred = "pred", col_obs = "target1", bins = 60, ...)

Arguments

data	dataframE
col_title	character vector denoting title column, Default: 'title'
col_pred	character vector denoting column with predictions, Default: 'preds'
col_obs	character vecor denoting column with observed values, Default: 'target1'
bins	number of bins used for histograms, Default: 60
...	additional arguments passed to the facet_wrap function of the histogramss

Value

list with two plots

See Also

UQ brewer.pal

Examples

  form = as.formula( 'displacement~cylinders+mpg')

  df = ISLR::Auto %>%
    pipelearner::pipelearner() %>%
    pipelearner::learn_models( rpart::rpart, form ) %>%
    pipelearner::learn_models( randomForest::randomForest, form ) %>%
    pipelearner::learn_models( e1071::svm, form ) %>%
    pipelearner::learn() %>%
    f_predict_pl_regression( 'name' ) %>%
    unnest(preds)

  f_predict_plot_regression_distribution(df
                                         , col_title = 'model'
                                         , col_pred = 'pred'
                                         , col_obs = 'target1')

erblast/oetteR documentation built on May 27, 2019, 12:11 p.m.