How Sure Is It?

A prediction on its own hides how close the vote was. You want the model's confidence too.

Task: write top_confidence(flowers, k).

• flowers is a list of iris flowers, each 4 numbers in the iris column order.
• Load iris with load_iris(as_frame=True) and split with train_test_split(..., test_size=0.25, random_state=0).
• Train a KNeighborsClassifier(n_neighbors=k) on the training rows.
• For each flower, find the highest of its predicted probabilities, which is the probability of the species the model would predict.
• Return those values as a list, each rounded to 2 decimal places, in the same order as flowers.
• Pass the flowers as a DataFrame with the iris column names.