Skip to contents

Runs the predictive models for the given datasets.

Usage

run_models(
  datasets,
  models,
  seed = sample(1:2^15, 1),
  training_size = 0.7,
  print_errors = FALSE,
  metrics = mldash::get_all_metrics(),
  save_model_fits = FALSE,
  timeout = 60
)

Arguments

datasets

the datasets to run the models with. Results from read_ml_datasets().

models

the models to run. Results from read_ml_models().

seed

random seed to set before randomly selecting the training dataset.

training_size

the proportion of the data that should be used for training. The remaining percentage will be used for validation.

print_errors

if TRUE errors will be printed while the function runs. Errors will always be saved in the returned object.

metrics

list of model performance metrics from the yardstick package. See https://yardstick.tidymodels.org/articles/metric-types.html for more information.

timeout

the maximum amount of time (in seconds) a model is allowed to run before it is interrupted, can be Inf to never expire. Note that not all models can be interrupted. See R.utils::withTimeout() for more information.

Value

a data.frame with the results of all the models run against all the datasets.