emm.threshold package¶
Submodules¶
emm.threshold.threshold_decision module¶
- emm.threshold.threshold_decision.decide_threshold(dataset_scored, aggregation_layer=False)¶
Get threshold decision curves
- Args:
dataset_scored: dataset from train_test_model(), with valid column. aggregation_layer: use aggregation layer? default is False.
- Returns:
dictionary with threshold decision curves
- Parameters:
aggregation_layer (
bool)
- emm.threshold.threshold_decision.get_threshold_curves_parameters(best_candidate_df, score_col='nm_score', aggregation_layer=False, aggregation_method='name_clustering', positive_set_col='positive_set')¶
Get threshold decision curves
- Args:
best_candidate_df: dataframe with the best candidates score_col: which score column to use, default is ‘nm_score’. For aggregation use ‘agg_score’. aggregation_layer: use aggregation layer? default is False. aggregation_method: which aggregation method is used? ‘name_clustering’ or ‘mean_score’. positive_set_col: name of positive set column in best candidates df. default is ‘positive_set’
- Returns:
dictionary with threshold decision curves
- Parameters:
score_col (
str)aggregation_layer (
bool)aggregation_method (
str)positive_set_col (
str)
- Return type:
dict