HistGradientBoostingClassifier copes with missing values by itself, with no imputer in front. You will punch holes in the iris flower data and check it still works.
Write score_with_gaps(column, every).
load_iris(return_X_y=True, as_frame=True) and split it with train_test_split(X, y, test_size=0.25, random_state=0, stratify=y).X_train and X_test. In both copies, set every every-th row of the column named column to np.nan, starting from the first row. Use .iloc positions, so the rows chosen are positions 0, every, 2 * every and so on in each table.HistGradientBoostingClassifier(random_state=0) on the training copy and score it on the test copy.Return the test score, rounded to 3 places.