Using python, implement your own version of the k-means algorithm using both mean and median. (We will focus on K-Medoids next week.) Create your implementation that takes: a dataset x, a list of center points c (each of n-dimension) and a enum type that indicates if you are using mean or median. (You can try to pass a function but you will have other dependent code that is based on whether you using the mean or median.) For the sample data set use euclidean distance and do not terminate your algorithm until none of the center points have changed. Use the following data: x = [[1, 2], [2, 1], [1, 3], [5, 4], [6, 3], [7, 2], [6, 1]] Q1: k-means, use center points: c = [[2, 2], [3, 4], [6, 21] Q2: k-means, use center points: c = [[3, 3], [4,1]] Q3: k-medians, use center points: c = [[2, 2], [3, 4], [6, 211 For each problem show the iterations, calculating the new center points (2-dimensional center points). Output the list of cluster values that each datapoint in x was assigned. (Cluster #'s should start at 0.)