R/feature_importance.R
find_permuted_perf_metric.Rd
Requires the future.apply
package
find_permuted_perf_metric( test_data, trained_model, outcome_colname, perf_metric_function, perf_metric_name, class_probs, feat, seed )
test_data | held out test data: dataframe of outcome and features |
---|---|
trained_model | trained model from caret |
outcome_colname | Column name as a string of the outcome variable (default |
perf_metric_function | Function to calculate the performance metric to be used for cross-validation and test performance. Some functions are provided by caret (see defaultSummary). Defaults: binary classification = |
perf_metric_name | The column name from the output of the function provided to perf_metric_function that is to be used as the performance metric. Defaults: binary classification = |
class_probs | whether to use class probabilities |
feat | feature or group of correlated features to permute |
seed | Random seed (default: |
vector of mean permuted auc and mean difference between test and permuted auc
Begüm Topçuoğlu, topcuoglu.begum@gmail.com
Zena Lapp, zenalapp@umich.edu