briverse17

Keybase proof

I hereby claim:

• I am briverse17 on github.
• I am briverse17 (https://keybase.io/briverse17) on keybase.
• I have a public key ASClqICkmZWZzzb0EGXAf22qkhszbwGm45mQ6FchnmLmJgo

To claim this, I am signing this object:

briverse17

def isConverged(self, new_medoids):
        return set([tuple(x) for x in self.medoids]) == set([tuple(x) for x in new_medoids])

briverse17

new_medoids = []
        for i in range(0, self.k):
            new_medoid = self.medoids[i]
            old_medoids_cost = self.medoids_cost[i]
            for j in range(len(clusters[i])):
                
                #Cost of the current data points to be compared with the current optimal cost
                cur_medoids_cost = 0
                for dpoint_index in range(len(clusters[i])):
                    cur_medoids_cost += euclideanDistance(clusters[i][j], clusters[i][dpoint_index])

briverse17

for i in range(self.max_iter):
            #Labels for this iteration
            cur_labels = []
            for medoid in range(0,self.k):
                #Dissimilarity cost of the current cluster
                self.medoids_cost[medoid] = 0
                for k in range(len(X)):
                    #Distances from a data point to each of the medoids
                    d_list = []                    
                    for j in range(0,self.k):

briverse17

def initMedoids(self, X):
        ''' 
        Parameters
        ----------
        X: input data. 
        '''
        self.medoids = []
        
        #Starting medoids will be random members from data set X
        indexes = np.random.randint(0, len(X)-1,self.k)

briverse17

class k_medoids:
    def __init__(self, k = 2, max_iter = 300, has_converged = False):
        ''' 
        Class constructor
        Parameters
        ----------
        - k: number of clusters. 
        - max_iter: number of times centroids will move
        - has_converged: to check if the algorithm stop or not
        '''

briverse17

class k_medoids:
    def __init__(self, k = 2, max_iter = 300, has_converged = False):
        ''' 
        Class constructor
        Parameters
        ----------
        - k: number of clusters. 
        - max_iter: number of times centroids will move
        - has_converged: to check if the algorithm stop or not
        '''