Section 6.4 ActiveCode Challenge: Decision Tree Classifier for Obstacle Sorting

In this exercise, you will build a pure Python decision tree classification loop. Standard Runestone CodeLens and ActiveCode environments do not support external libraries like scikit-learn or opencv. Therefore, you will write pure Python logic that mimics a decision tree classifier to sort data coordinates into "obstacle" or "free space".

Subsection 6.4.1 Section 6.4 Interactive Exercises

Subsubsection 6.4.1.1 Exercise 6.4.1: Decision Tree Classifier Challenge

Complete the decision tree classifier below to sort incoming coordinate telemetry into "OBSTACLE" vs. "FREE_SPACE".

Task:

Complete the predict_decision_tree function according to these decision rules:
- If x_coord < 2.0 AND y_coord < 2.0, classify as "OBSTACLE".
- Otherwise, classify as "FREE_SPACE".
🔗
🔗
Run the script to evaluate the accuracy of your classifier against the test dataset!

def predict_decision_tree(x_coord, y_coord):
    # TODO: Implement decision tree logic
    # Rule: If x < 2.0 AND y < 2.0 -> return "OBSTACLE"
    # Otherwise -> return "FREE_SPACE"

predicted_label = "FREE_SPACE"
    return predicted_label

# Telemetry Test Dataset: List of ((x, y), ground_truth_label)
test_dataset = [
    ((0.5, 1.2), "OBSTACLE"),
    ((3.1, 0.8), "FREE_SPACE"),
    ((1.8, 1.9), "OBSTACLE"),
    ((2.5, 4.0), "FREE_SPACE"),
    ((0.2, 3.5), "FREE_SPACE")
]

print("--- Running Decision Tree Classifier Evaluation ---")
correct_predictions = 0

for i, (coords, true_label) in enumerate(test_dataset):
    x, y = coords
    pred_label = predict_decision_tree(x, y)

is_correct = (pred_label == true_label)
    if is_correct:
        correct_predictions += 1

print(f"Sample {i+1}: Coords=({x}, {y}) | True={true_label:10s} | Pred={pred_label:10s} | Match={is_correct}")

accuracy = (correct_predictions / len(test_dataset)) * 100.0
print(f"\nClassifier Accuracy: {accuracy:.1f}%")

You have attempted of activities on this page.

Prev Top Next