Scribd
Upload
16 views

K Means Algorithm

The document discusses implementing K-means clustering using Python. It describes the algorithm steps which include initializing centroid values, assigning clusters, updating centroids, and predicting clusters. It also includes sample Python code for a K-means class with methods for these steps.

Uploaded by

kishore5783
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
16 views

K Means Algorithm

The document discusses implementing K-means clustering using Python. It describes the algorithm steps which include initializing centroid values, assigning clusters, updating centroids, and predicting clusters. It also includes sample Python code for a K-means class with methods for these steps.

Uploaded by

kishore5783
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

K-MEAN’S CLUSTERING

EX. NO: 04

AIM:
To implement the k-means clustering using python.

ALGORITHM:
Step 1: start the code cell.
Step 2: to assign cluster value to database.
Step 3: using SQL, Read the mean value to data frame.
Step 4: create SQL command to training and prediction
Step 5: Calculating error for K values to database.
Step 6: initialize the k mean value to clustering datas.
Step 7: python code to demonstrate the k mean data.
Step 8: store all the data in the variable.
Step 9: Stop.

SOURCE CODE:
class K-means:

k: int , number of clusters

seed: int, will be randomly set if None


max_iter: int, number of iterations to run algorithm, default: 200

centroids: array, k, number_features

def __init__(self, k, seed = None, max_iter = 200):


self.k = k
self.seed = seed
if self.seed is not None:
np.random.seed(self.seed)
self.max_iter = max_iter

def initialise_centroids(self, data):

initial_centroids = np.random.permutation(data.shape[0])[:self.k]
self.centroids = data[initial_centroids]
return self.centroids
def assign_clusters(self, data):

data: array or matrix, number_rows, number_features

if data.ndim == 1:
data = data.reshape(-1, 1)
dist_to_centroid = pairwise_distances(data, self.centroids, metric = 'euclidean')
self.cluster_labels = np.argmin(dist_to_centroid, axis = 1)
return self.cluster_labels

def update_centroids(self, data):


data: array or matrix, number_rows, number_features

self.centroids = np.array([data[self.cluster_labels == i].mean(axis = 0) for i in


range(self.k)])
return self.centroids

def predict(self, data):

data: array or matrix, number_rows, number_features

return self.assign_clusters(data)

self.centroids = self.initialise_centroids(data)

self.cluster_labels = self.assign_clusters(data)
self.centroids = self.update_centroids(data)
if iter % 100 == 0:
print("Running Model Iteration %d " %iter)
print("Model finished running")
return self

Output:
VAR1 and VAR2 predict a classification
VAR1 VAR2 CLASS
0.307 0.654 1
0.089 0.528 1
0.606 0.258 1
RESULT:

Thus the python program for K-mean’s clustering was successfully executed and the
output was verified.

You might also like