Format metrics for plotting — format_metrics • BirdFlowPipeline

This formats the metris

Usage

format_metrics(
  train_metrics,
  test_metrics,
  cor_threshold,
  cor_metric,
  this_model = NULL,
  include = NULL,
  use = "test",
  long = TRUE
)

Arguments

train_metrics, test_metrics: Similarly formatted evaluation metrics for all the models in the grid search. This is saved to an Rds during Yangkang's grid search workflow and is calculated on either the training models or testing models.
cor_threshold: A threshold in correlation between model distributions and training (S&T) distributions below which models will be excluded for selection by some of the criteria.
cor_metric: The particular correlation metric to which the threshold is applied - a column in metrics.
this_model: The model id (in metrics$model) for the current model. This is used when making a model report to indicate which model the report is for,
include: list of metric columns to include in the output.
use: either "test" or "train" indicating which metrics should be plotted.
long: if TRUE use long format with metric and value column, otherwise use wider format with a column for each metric.

Value

A modified version of metrics with models labeled by type. Will include both the best and candidate models. Columns:

model: the model file name - used as unique ID
type: Classified model types. One of the types defined by identify_best_models() or one of this, candidate or low_cor
metric: one of the metric names - a column in the input
value: the metric value.