immuneML.ml_metrics package¶

Submodules¶

immuneML.ml_metrics.ClassificationMetric module¶

class immuneML.ml_metrics.ClassificationMetric.ClassificationMetric(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)[source]¶

Bases: Enum

ACCURACY = 'accuracy_score'¶

AUC = 'roc_auc_score'¶

BALANCED_ACCURACY = 'balanced_accuracy_score'¶

CONFUSION_MATRIX = 'confusion_matrix'¶

F1_MACRO = 'f1_score_macro'¶

F1_MICRO = 'f1_score_micro'¶

F1_WEIGHTED = 'f1_score_weighted'¶

LOG_LOSS = 'log_loss'¶

PRECISION = 'precision_score'¶

RECALL = 'recall_score'¶

static get_metric(metric_name: str)[source]¶

static get_probability_based_metric_types()[source]¶

static get_search_criterion(metric)[source]¶

static get_sklearn_score_name(metric)[source]¶

immuneML.ml_metrics.ClusteringMetric module¶

immuneML.ml_metrics.ClusteringMetric.is_external(metric: str)[source]¶

immuneML.ml_metrics.ClusteringMetric.is_internal(metric: str)[source]¶

immuneML.ml_metrics.MetricUtil module¶

class immuneML.ml_metrics.MetricUtil.MetricUtil[source]¶

Bases: object

static get_metric_fn(metric: ClassificationMetric)[source]¶

static score_for_metric(metric: ClassificationMetric, predicted_y, predicted_proba_y, true_y, classes)[source]¶: Note: when providing label classes, make sure the ‘positive class’ is sorted last. This sorting should be done automatically when accessing Label.values

immuneML.ml_metrics.ml_metrics module¶

immuneML.ml_metrics.ml_metrics.f1_score_macro(true_y, predicted_y, sample_weight=None)[source]¶

immuneML.ml_metrics.ml_metrics.f1_score_micro(true_y, predicted_y, sample_weight=None)[source]¶

immuneML.ml_metrics.ml_metrics.f1_score_weighted(true_y, predicted_y, sample_weight=None)[source]¶

immuneML.ml_metrics.ml_metrics.roc_auc_score(true_y, predicted_y, sample_weight=None, labels=None)[source]¶