Getting invalid syntax error on line 26 if xi_distance > d_max:

import copy
import math

class obj_kMeans:
    
    def __init__(self,init_centroid,k):
        self.k = k
        self.clusters = None
        self.centroids = np.empty((self.k,2))
        self.initial_centroid = init_centroid
    
    def distance(self,x,y):
         return np.linalg.norm(y-x, ord=2)
        
    def init_centroids(self, x):
        centroids = self.centroids
        centroids[0] = self.initial_centroid
        for i in range(1, self.k):
            d_max = 0
            d_max_cen = centroids[0]
            for x_i in x:
                if x_i in centroids:
                    continue
                xi_distance = sum([self.distance(x_i, prev) for prev in centroids]
                if xi_distance > d_max:
                    d_max = xi_distance
                    d_max_cen = x_i
                centroids[i] = d_max_cen
        return centroids
    def cost_function(self,clusters):
         squared_errors = list()
         for i in range(self.k):
             points = clusters[i][1]
             centroid = clusters[i][0]
             errors = [self.distance(centroid,point)**2 for point in points]
             total_error = sum(errors)
             squared_errors.append(total_error)
         return sum(squared_errors)
    def cluster(self,x):
        centroids = self.init_centroids(x)
        clusters = {}
        for i in range(self.k):
            cluster_list = (centroids[i],list())
            clusters[i] = cluster_list
        itr = 0
        while True:
             for point in x:
                 dist_min = self.distance(centroids[0],point)
                 num = 0
                 cls_point = 0
                 for centroid in centroids:
                     d = self.distance(centroid,point)
                     if d < dist_min:
                         cls_point = num
                         dist_min = d
                     num +=1
                 clusters[cls_point][1].append(point)
         self.clusters = copy.deepcopy(clusters)
         centroids = list()
         converges = True
         for i in range(self.k):
             dp = clusters[i][1]
             dp.append(clusters[i][0])
             centroid_p = np.mean(np.array(dp), axis=0)
             if not ((centroid_p == clusters[i][0]).all()):
                 converges = False
             centroids.append(centroid_p)
             cluster_list = (centroid_p,list())
             clusters[i] = cluster_list
        itr +=1
        if converges:
            break
        self.centroids = centroids


kMeans_implement1 = obj_kMeans(k=k1, init_centroid = i_point1)
kMeans_implement2 = obj_kMeans(k=k2, init_centroid = i_point2)

kMeans_implement1.cluster(data)
kMeans_implement2.cluster(data)

cost_val1 = kMeans_implement1.cost_function(kMeans_implement1.clusters)
cost_val2 = kMeans_implement2.cost_function(kMeans_implement2.clusters)
                                  
print("centroid 1", kMeans_implement1.centroids)
print("centroid 2", kMeans_implement2.centroids)

print("cost 1", cost_val1)
print("cost 2", cost_val2)