Implementation:Scikit learn contrib Imbalanced learn sensitivity specificity support

Knowledge Sources	imbalanced-learn imbalanced-learn Docs
Domains	Machine_Learning, Model_Evaluation, Imbalanced_Learning
Last Updated	2026-02-09 03:00 GMT

Overview

Concrete tool for computing per-class sensitivity, specificity, and support provided by the imbalanced-learn library.

Description

The sensitivity_specificity_support function computes sensitivity (recall), specificity (true negative rate), and support (number of true instances) for each class. It also provides sensitivity_score and specificity_score convenience wrappers. Supports binary, multiclass, macro, weighted, and micro averaging.

Usage

Import this function (or the convenience wrappers sensitivity_score and specificity_score) to evaluate per-class or averaged sensitivity and specificity.

Code Reference

Source Location

• Repository: imbalanced-learn
• File: imblearn/metrics/_classification.py
• Lines: L48-530 (sensitivity_specificity_support: L48-314, sensitivity_score: L315-414, specificity_score: L431-530)

Signature

def sensitivity_specificity_support(
    y_true,
    y_pred,
    *,
    labels=None,
    pos_label=1,
    average=None,
    warn_for=("sensitivity", "specificity"),
    sample_weight=None,
    zero_division="warn",
):
    """Returns (sensitivity, specificity, support) tuple."""

def sensitivity_score(
    y_true, y_pred, *, labels=None, pos_label=1, average="binary", sample_weight=None
):
    """Returns sensitivity (recall) score."""

def specificity_score(
    y_true, y_pred, *, labels=None, pos_label=1, average="binary", sample_weight=None
):
    """Returns specificity (true negative rate) score."""

Import

from imblearn.metrics import sensitivity_score, specificity_score
from imblearn.metrics import sensitivity_specificity_support

I/O Contract

Inputs

Name	Type	Required	Description
y_true	array-like of shape (n_samples,)	Yes	Ground truth labels
y_pred	array-like of shape (n_samples,)	Yes	Predicted labels
average	str or None	No	Averaging: 'binary', 'macro', 'weighted', 'micro', None
pos_label	int	No	Positive class for binary (default: 1)

Outputs

Name	Type	Description
sensitivity	float or ndarray	Sensitivity (recall) per class or averaged
specificity	float or ndarray	Specificity per class or averaged
support	ndarray	Number of true instances per class

Usage Examples

from imblearn.metrics import sensitivity_score, specificity_score

y_true = [0, 1, 2, 0, 1, 2]
y_pred = [0, 2, 1, 0, 0, 1]

sens = sensitivity_score(y_true, y_pred, average="macro")
spec = specificity_score(y_true, y_pred, average="macro")
print(f"Sensitivity: {sens:.3f}, Specificity: {spec:.3f}")

# Per-class
from imblearn.metrics import sensitivity_specificity_support
sen, spe, sup = sensitivity_specificity_support(y_true, y_pred)
print(f"Per-class sensitivity: {sen}")
print(f"Per-class specificity: {spe}")
print(f"Support: {sup}")

Related Pages

Implements Principle

• Principle:Scikit_learn_contrib_Imbalanced_learn_Sensitivity_Specificity_Analysis

Requires Environment

• Environment:Scikit_learn_contrib_Imbalanced_learn_Python_Scikit_learn