Neighbours or a Tree?

You have two models and one dataset. Which one should you trust? Give both the same fair test.

Task: write compare_models(name, seed). name is "iris" or "wine".

• Load the dataset as X and y, and split with train_test_split(X, y, test_size=0.25, random_state=seed, stratify=y).
• Train a KNeighborsClassifier(n_neighbors=5) and a DecisionTreeClassifier(random_state=0) on the same training rows.
• Score both on the same test rows. Round each score to 3 decimal places.
• Return a dict: {"knn": knn score, "tree": tree score, "winner": ...}.
• "winner" is "knn" or "tree" for whichever rounded score is higher, or "tie" if the rounded scores are equal.