Calculating manhattan distance manually
While euclidean distance is very popular, it only scales well beyond two or three-dimensional data. In these cases, you can use manhattan distance as an alternative. It has the advantage of working exceptionally well with datasets with many categorical features.
Practice calculating it manually with NumPy, which has been loaded under its standard alias np.
Este ejercicio forma parte del curso
Anomaly Detection in Python
Instrucciones del ejercicio
- Find the absolute differences between the elements of
MandN. - Find the sum of differences to calculate the final manhattan distance.
Ejercicio interactivo práctico
Prueba este ejercicio completando el código de muestra.
M = np.array([14, 17, 18, 20, 14, 12, 19, 13, 17, 20])
N = np.array([63, 74, 76, 72, 64, 75, 75, 61, 50, 53])
# Subtract M from N and find the absolute value
abs_diffs = ____
# Calculate the final manhattan distance
manhattan_dist_MN = ____
print(manhattan_dist_MN)