Plot annotated confusion matrix for binary classifier

annotated_confusion_matrix.py

import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix
import numpy as np

def plot_confusion_matrix(cf_matrix, target_names=None):
  """Plot annotated confusion matrix with absolute values and percentages
     Args: 
       cf_matrix: ndarray of shape (2, 2) Confusion Matrix 
  """
    group_names = ['TN', 'FP', 'FN', 'TP']
    group_counts = ["{0:.0f}".format(value) for value in 
                    cf_matrix.flatten()]
    group_percentages = ["{0:.2%}".format(value) for value in 
                         cf_matrix.flatten()/np.sum(cf_matrix)]
    labels = [f"{v1}\n{v2}\n{v3}" for v1, v2, v3 in 
              zip(group_names,
                  group_counts,
                  group_percentages)
             ]
    labels = np.asarray(labels).reshape(2, 2)

    plt.figure(figsize = (8, 6))
    sns.heatmap(cf_matrix, annot=labels, fmt='', cmap='Blues',  annot_kws={"size": 14})
    if target_names:
        tick_marks = range(len(target_names))
        plt.xticks(tick_marks, target_names)
        plt.yticks(tick_marks, target_names)

    precision = cf_matrix[1, 1] / sum(cf_matrix[:, 1])
    recall    = cf_matrix[1, 1] / sum(cf_matrix[1,:])
    accuracy  = np.trace(cf_matrix) / float(np.sum(cf_matrix))
    f1_score  = 2 * precision * recall / (precision + recall)
    stats_text = "Precision.={:0.3f}\nRecall{:0.3f}\n\nAccuracy={:0.3f}\nF1 Score={:0.3f}".format(
        precision, recall, accuracy, f1_score)
    plt.xlabel('Predicted label {}'.format(stats_text))
    plt.ylabel("True Label")
    plt.show()

conf_matrix = confusion_matrix(y_label, pred_label)
plot_confusion_matrix(conf_matrix, target_names=['c1', 'c2'])