This is for a class and I would really appreciate your help! I made some changes based on a comment I received, but now I get another error.. I need to modify an existing function that implements the mean-shift algorithm, but instead of initializing all the points as the first set of centroids, the function creates a grid of centroids with the grid based on the radius. I also need to delete the centroids that don't contain any data points. My issue is that I don't understand how to fix the error I get!
---------------------------------------------------------------------------
IndexError Traceback (most recent call last)
<ipython-input-7-de18ffed728f> in <module>()
49 centroids = initialize_centroids(x)
50
---> 51 new_centroids = update_centroids(x, centroids, r = 1)
52
53 print(len(centroids))
<ipython-input-7-de18ffed728f> in update_centroids(data, centroids, r)
26 #print(len(centroids))
27 #print(range(len(centroids)))
---> 28 centroid = centroids[i]
29 for data_point in data:
30 if np.linalg.norm(data_point - centroid) < r:
IndexError: index 2 is out of bounds for axis 0 with size 2
I tried using the range of the input dataset as boundaries for a grid, with the points separated by the radius.
from sklearn import datasets
import numpy as np
import matplotlib.pyplot as plt
def initialize_centroids(data, r = 1):
'''Creates a grid of centroids with grid based on radius'''
data = np.array(data)
xi,yi = min(range(len(data))), max(range(len(data)))
mx = np.arange(xi,yi,r)
x,y = np.meshgrid(mx,mx)
centroids=np.vstack([x.ravel(), y.ravel()])
return centroids
#update centroids based on mean of points that fall within a specified radius of each centroid
def update_centroids(data, centroids, r = 1):
new_centroids = []
for i in centroids:
in_radius = []
centroid = centroids[i] #this is where the error occurs
for data_point in data:
if np.linalg.norm(data_point - centroid) < radius:
in_radius.append(data_point) #this list is appended by adding the new centroid to it if the above conition is satisfied.
new_centroid = np.mean(in_radius, axis=0)
#maybe another way to do the next part
new_centroids.append(tuple(new_centroid))
unique_centroids = sorted(list(set(new_centroids))) #for element in in_radius, if element in set skip else set.append(element(in_rad)). append does not work with set.
new_centroids = {i:np.array(unique_centroids[i]) for i in range(len(unique_centroids))}
return new_centroids
#test function on:
x, y = datasets.make_blobs(n_samples=300, n_features = 2, centers=[[0, 7], [0, -7], [5,7], [5, 0]])
centroids = initialize_centroids(x)
new_centroids = update_centroids(x, centroids, radius = 2)
print(len(centroids))
print()
print(len(new_centroids))
#code for plotting initially:
plt.scatter(x[:,0], x[:,1], color = 'k')
for i in range(len(new_centroids)):
plt.scatter(new_centroids[i][0], new_centroids[i][1], s=200, color = 'r', marker = "*")
#code for plotting updated centroids:
new_centroids = update_centroids(x, new_centroids, radius = 2)
plt.scatter(x[:,0], x[:,1], color = 'k')
for i in range(len(new_centroids)):
plt.scatter(new_centroids[i][0], new_centroids[i][1], s=200, color = 'r', marker = "*")
#code for iterations:
def iterate_to_conv(data, max_iter=100):
centroids = initialize_centroids(data)
iter_count = 0
while iter_count <= max_iter:
new_centroids = update_centroids(data, centroids, radius = 2)
centroids = new_centroids
iter_count += 1
return centroids
centroids = iterate_to_conv(x)
plt.scatter(x[:,0], x[:,1], color = 'k')
for i in range(len(centroids)):
plt.scatter(centroids[i][0], centroids[i][1], s=200, color = 'r', marker = "*")
The function needs to return the number of final centroids. I haven't gotten ahead far enough to know how the entire implementation of mean-shift would work with this function..
When you are running that loop: for i in centroids the i that is iterated through centroids isn't a number, it is a vector which is why an error is pops up. For example, the first i value might be equal to [0 1 2 0 1 2 0 1 2]. So to take an index of that doesn't make sense. What your code is saying to do is to take centroid = centroid[n1 n2 nk]. To fix it, you really need to change how your initialize centroid function works. Meshgrid also won't create an N dimensional grid, so your meshgrid might work for 2 dimensions but not N. I hope that helps.