Classifier metrics

clf_metrics.py

import matplotlib.pyplot as plt
from sklearn.metrics import (
    PrecisionRecallDisplay, precision_recall_curve, RocCurveDisplay, roc_curve
    )

p1, r1, thresh1 = precision_recall_curve(yactual, ypred1)
p2, r2, thresh2 = precision_recall_curve(yactual, ypred2)

# Determine threshold that maximizes f1 score. 
f1_1 = 2 * (p1 * r1) / (p1 + r1)
f1_2 = 2 * (p2 * r2) / (p2 + r2)

best_f1_1 = np.argmax(f1_1)
best_f1_2 = np.argmax(f1_2)

best_thresh1 = thresh1[best_f1_1]
best_thresh2 = thresh1[best_f1_2]

print(f"best_thresh1: {best_thresh1:.5f}")
print(f"best_thresh2: {best_thresh2:.5f}")

fig, (ax1, ax2) =  plt.subplots(1, 2, figsize=(10, 4.5), tight_layout=True)

pr_disp1 = PrecisionRecallDisplay.from_predictions(
    yactual, ypred1, name="mm", plot_chance_level=False, ax=ax1, color="#CD0066"
    )
ax1.grid(True)
pr_disp1.ax_.set_title("mm Precision-Recall curve", fontsize=9)
roc_disp1 = RocCurveDisplay.from_predictions(
    yactual, ypred1, name="mm", plot_chance_level=True, ax=ax2, color="#191964"
    )
ax2.grid(True)
roc_disp1.ax_.set_title("mm ROC curve", fontsize=9)

plt.show()