Matriz de confusión

• Es una matriz que permite visualizar el desempeño de un clasificador.

• La organización típica es la presentada a continuación:

         | Pronóstico
         |  PP    PN
---------|------------
      P  |  TP    FN
Real     |
      N  |  FP    TN


 P - Positive                 TP - Verdadero positivo (correcto)
 N - Negative                 TN - Verdadero negativo (correcto)
PP - Predicted Positive       FN - Falso negativo (mal clasificado)
PN - Predicted Negative       FP - Falso positivo (mal clasificado)

#
# Cálculo a partir de los valores reales y los pronósticos
#
from sklearn.metrics import confusion_matrix

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

#
#         | Pronostico
#         |   0    1
# --------|-----------
#       0 |  10    1
# Real    |
#       1 |   3    1
#

confusion_matrix(
    # -------------------------------------------------------------------------
    # Ground truth (correct) target values.
    y_true=y_true,
    # -------------------------------------------------------------------------
    # Estimated targets as returned by a classifier.
    y_pred=y_pred,
    # -------------------------------------------------------------------------
    # List of labels to index the matrix.
    labels=None,
    # -------------------------------------------------------------------------
    # Normalizes confusion matrix over the true (rows), predicted (columns)
    # conditions or all the population.
    # 'true', 'pred', 'all'
    normalize=None,
)

array([[10,  1],
       [ 3,  1]])

confusion_matrix(
    y_true=y_true,
    y_pred=y_pred,
    labels=[1, 0],
    normalize=None,
)

array([[ 1,  3],
       [ 1, 10]])

import pandas as pd

pd.DataFrame(
    confusion_matrix(
        y_true=y_true,
        y_pred=y_pred,
        labels=[1, 0],
        normalize=None,
    ),
    columns=["PP=1", "PF=0"],
    index=["P=1", "F=0"],
)

#
#          | Pronóstico
#          |  PP    PN
# ---------|------------
#       P  |  TP    FN
# Real     |
#       N  |  FP    TN
#

	PP=1	PF=0
P=1	1	3
F=0	1	10

#
# Extracción de los elementos de la matriz de confusión
#
tn, fp, fn, tp = confusion_matrix(
    y_true=y_true,
    y_pred=y_pred,
).ravel()

display(
    tn,
    fp,
    fn,
    tp,
)

10

1

3

1

Matriz de confusión para más de dos clases

import matplotlib.pyplot as plt
import numpy as np
from sklearn import datasets, svm
from sklearn.metrics import ConfusionMatrixDisplay
from sklearn.model_selection import train_test_split

iris = datasets.load_iris()
X = iris.data
y = iris.target
class_names = iris.target_names


X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)

classifier = svm.SVC(kernel="linear", C=0.01).fit(X_train, y_train)

np.set_printoptions(precision=2)

disp = ConfusionMatrixDisplay.from_estimator(
    classifier,
    X_test,
    y_test,
    display_labels=class_names,
    cmap=plt.cm.Blues,
    normalize=None,
)
print(disp.confusion_matrix)

plt.show()

[[13  0  0]
 [ 0 10  6]
 [ 0  0  9]]

Normalización

#
# Normalización sobre las filas (true)
#
ConfusionMatrixDisplay.from_estimator(
    classifier,
    X_test,
    y_test,
    display_labels=class_names,
    cmap=plt.cm.Blues,
    normalize="true",
)
plt.show()

#
# Normalización sobre las columnas (pred)
#
ConfusionMatrixDisplay.from_estimator(
    classifier,
    X_test,
    y_test,
    display_labels=class_names,
    cmap=plt.cm.Blues,
    normalize="pred",
)
plt.show()

#
# Normalización sobre todos los datos
#
ConfusionMatrixDisplay.from_estimator(
    classifier,
    X_test,
    y_test,
    display_labels=class_names,
    cmap=plt.cm.Blues,
    normalize="all",
)
plt.show()