overview_metrics: An overview of the results with all available metrics

Description Usage Arguments Value

Description

overview_metrics use the result of the tuning and diagnoses it using a list of selected metric. It's useful to see how other metrics evaluate the result tuned with only one given metric which is subject to biases.

Usage

1
2
3
overview_metrics(method_result, internal_metrics = NULL,
  true_partition = NULL, external_metrics = NULL, print = T,
  plot = T)

Arguments

method_result

a list returned by built-in methods i.e. a list with an element partition and data_returned. To use directly the result of the function tuning, use the element $method_res of the result of the tuning.

internal_metrics

a character vector indicating the name of internal metrics to be used. Be careful, the internal metric used must correspond to the data returned by the selected method (a distance matrix? an affinity matrix?, a feature matrix? ...)

true_partition

factor or integer vector being a classification of the patients to be used with external metrics.

external_metrics

a character vector with the name of external metrics to be used with the provided ground-truth. If NULL, all the external metrics are evaluated.

print

logical. Print or not the values of metrics in the console.

plot

logical. Plot or not silhouette graph

Value

a data.frame with the value of metrics.


agapow/subtypr documentation built on May 5, 2019, 1:33 a.m.