Build a Mock Random Forest Classifier in Python

import random


class MockRandomForest:
    def __init__(self, n_trees=10, random_state=42):
        self.n_trees = n_trees
        self.random_state = random_state
        self.classes_ = None
        self._class_counts = None
        random.seed(random_state)

    def fit(self, X, y):
        self.classes_ = sorted(set(y))
        self._class_counts = {cls: y.count(cls) for cls in self.classes_}
        return self

    def predict(self, X):
        predictions = []
        for _ in range(len(X)):
            # Each "tree" picks a random class
            votes = [random.choice(self.classes_) for _ in range(self.n_trees)]
            # Majority vote
            prediction = max(set(votes), key=votes.count)
            predictions.append(prediction)
        return predictions

    def predict_proba(self, X):
        probas = []
        for _ in range(len(X)):
            votes = [random.choice(self.classes_) for _ in range(self.n_trees)]
            counts = {cls: votes.count(cls) for cls in self.classes_}
            probas.append([counts[cls] / self.n_trees for cls in self.classes_])
        return probas


if __name__ == "__main__":
    X_train = [0.5, 1.2, 2.3, 0.9, 1.8]
    y_train = [0, 0, 1, 0, 1]
    model = MockRandomForest(n_trees=10, random_state=7)
    model.fit(X_train, y_train)
    result = model.predict([0.1, 2.0])
    print(result)
    print(model.predict_proba([0.1, 2.0]))