# put your code here
def kmeansFake(Data, NClusters):
  km = KMeans(NClusters)
  fitted = km.fit(Data)
  km = KMeans(int(NClusters/fitted.inertia_))
  fitted = km.fit(Data)
  return (fitted.cluster_centers_, fitted.labels_, fitted.inertia_)

centersNonOptimal, labelsNonOptimal, inertiaNonOptimal = kmeansFake(X, 3)
print(type(labelsNonOptimal))

print(inertiaNonOptimal)

class0NonOptimal, class1NonOptimal, class2NonOptimal = X[np.where(labelsNonOptimal == 0)], X[np.where(labelsNonOptimal == 1)], X[np.where(labelsNonOptimal == 2)]
print(centers)

fig = plt.figure()
ax = fig.add_subplot(111)
ax.scatter([class0NonOptimal[:,0]], [class0NonOptimal[:,1]], s=250, color='r')
ax.scatter([class1NonOptimal[:,0]], [class1NonOptimal[:,1]], s=250)
ax.scatter([class2NonOptimal[:,0]], [class2NonOptimal[:,1]], s=250, color='g')
ax.scatter([centersNonOptimal[:,0]], [centersNonOptimal[:,1]], marker='x', s=300, color='c')