import numpy as np def find_duplicate_records(arr: np.ndarray) -> np.ndarray: """Returns duplicate records from a numpy array Example: >>> arr = np.array([1, 2, 3, 2, 4, 5, 6, 3, 7, 8, 1, 9, 2, 3]) """ unique_elements, counts = np.unique(arr, return_counts=True) return unique_elements[counts > 1]