Threshopt

Automatic threshold optimization for binary classifiers.

Features

• Optimize decision thresholds based on any metric (e.g., accuracy, F1-score, G-Mean, Youden’s J)
• Supports cross-validation threshold optimization for robust model tuning
• Easy integration with any scikit-learn compatible model
• Built-in common metrics and ability to use custom metrics
• Visualize confusion matrices and prediction score distributions (optional)

Installation

pip install threshopt

Quickstart

from threshopt import optimize_threshold, gmean_score
from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split

X, y = make_classification(n_samples=100, n_features=5, random_state=42)
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)

model = RandomForestClassifier(random_state=42)
model.fit(X_train, y_train)

best_thresh, best_score = optimize_threshold(model, X_test, y_test, metric=gmean_score)
print(f"Best threshold: {best_thresh:.2f}, score: {best_score:.4f}")

License

Apache License 2.0

Links

• PyPI Project Page
• GitHub Repository