Machine Learning

Lab : 6
KMeans Clustering

2026
January 2025

12-162025
Perform Date: January 13-18, 2026

1 Objective
1.1 Implement K Means Clustering Algorithm on the given dataset

2 Description
K-Means Clustering is an unsupervised learning algorithm that is used to solve clustering prob-
lems in machine learning or data science. A cluster refers to a collection of data points aggregated
together because of certain similarities. We will define a target number k, which refers to the num-
ber of centroids you need in the dataset. A centroid is an imaginary or real location representing
the center of the cluster. Every data point is allocated to each of the clusters through reducing
the in-cluster sum of squares. In other words, the K-means algorithm identifies the k number of
centroids, and then allocates every data point to the nearest cluster, while keeping the centroids
as small as possible. The ‘means’ in the K-means refers to averaging of the data; that is, finding
the centroid.

3 Implementation Guidelines
3.1 Part A
[35]: # Imports
from [Link] import make_blobs
X, _ = make_blobs(n_samples=100, centers=3, n_features=2,
cluster_std=0.2, random_state=0)

[1]: # Scatter plot of the data points

import [Link] as plt
%matplotlib inline

[2]: # Using scikit-learn to perform K-Means clustering

from [Link] import KMeans
# Specify the number of clusters (3) and fit the data X

1
 Lab Manual: Machine Learning

[ ]: # Get the cluster centroids

[ ]: # Get the cluster labels

[4]: # Plotting the cluster centers and the data points on a 2D plane

[6]: # Calculate silhouette_score

[8]: # Import the KElbowVisualizer method

# Instantiate a scikit-learn K-Means model

# Instantiate the KElbowVisualizer with the number of clusters and the metric
# Fit the data and visualize

3.2 Part B
Hand Written Digit Recognition

[44]: import numpy as np # linear algebra

import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)

from [Link] import KMeans

from [Link] import load_digits
#digits dataset from scikit learn consists of 8x8 pixel images of digits

#Data plotting and visualization libraries

import [Link] as plt
import seaborn as sns

from [Link] import mode

from [Link] import accuracy_score, confusion_matrix

[45]: digits = load_digits() #load the dataset in digits

[45]: (1797, 64)

[29]: [Link]() #Dataset loaded is a dictionary

# data : flattened arrays/tensors used for clustering
# target : label associated with flattened array
#print([Link])

[29]: dict_keys(['data', 'target', 'frame', 'feature_names', 'target_names', 'images',

'DESCR'])

[30]: [Link][0:3] #flattened data for 3 images of the dataset

Computer Engineering Department, Faculty of Technology, D. D. University Page 2 of 5

 Lab Manual: Machine Learning

[30]: array([[ 0., 0., 5., 13., 9., 1., 0., 0., 0., 0., 13., 15., 10.,
15., 5., 0., 0., 3., 15., 2., 0., 11., 8., 0., 0., 4.,
12., 0., 0., 8., 8., 0., 0., 5., 8., 0., 0., 9., 8.,
0., 0., 4., 11., 0., 1., 12., 7., 0., 0., 2., 14., 5.,
10., 12., 0., 0., 0., 0., 6., 13., 10., 0., 0., 0.],
[ 0., 0., 0., 12., 13., 5., 0., 0., 0., 0., 0., 11., 16.,
9., 0., 0., 0., 0., 3., 15., 16., 6., 0., 0., 0., 7.,
15., 16., 16., 2., 0., 0., 0., 0., 1., 16., 16., 3., 0.,
0., 0., 0., 1., 16., 16., 6., 0., 0., 0., 0., 1., 16.,
16., 6., 0., 0., 0., 0., 0., 11., 16., 10., 0., 0.],
[ 0., 0., 0., 4., 15., 12., 0., 0., 0., 0., 3., 16., 15.,
14., 0., 0., 0., 0., 8., 13., 8., 16., 0., 0., 0., 0.,
1., 6., 15., 11., 0., 0., 0., 1., 8., 13., 15., 1., 0.,
0., 0., 9., 16., 16., 5., 0., 0., 0., 0., 3., 13., 16.,
16., 11., 5., 0., 0., 0., 0., 3., 11., 16., 9., 0.]])

[31]: #run KMeans clustering on [Link] for 1797 records and 64 features

[31]: (10, 64)

[32]: fig, ax = [Link](2, 5, figsize = (8,3)) #Create a figure and a set of␣
,→subplots( 2 rows and 5 columns)

centers = k_means.cluster_centers_.reshape(10,8,8)
#flattened image can't be viewed, re-transform/reshape/inverse transform it to␣
,→original form to view matrix shaped image

#reshape 10 rows of clusters (k_means.cluster_centers_ = 10,64) and 64 to 8 * 8␣

,→matrix

for axi, center in zip([Link], centers): #[Link]:flattening the image &␣

,→plotting relevant centers

[Link](xticks = [], yticks = [])

[Link](center, interpolation='nearest',cmap = [Link])␣
,→#imshow(matplotlib method) to render the image in notebook

Computer Engineering Department, Faculty of Technology, D. D. University Page 3 of 5

 Lab Manual: Machine Learning

[33]: labels = np.zeros_like(clusters) # blank labels

print(f"The labels are : {labels}")
print(f"\nThe size of labels is : {[Link]}")
print("The mask values are : ")
for i in range(10):
mask = (clusters == i)
#if a specific digit belongs to/equivalent a specific cluster then its True␣
,→else False

print(mask)
labels[mask] = mode([Link][mask])[0]

The labels are : [0 0 0 ... 0 0 0]

The size of labels is : (1797,)

The mask values are :
[False False False ... False False False]
[False False False ... False False False]
[False False False ... False False False]
[False False False ... False True True]
[False False False ... False False False]
[ True False False ... False False False]
[False False False ... False False False]
[False True True ... True False False]
[False False False ... False False False]
[False False False ... False False False]

[10]: #if a specific digit belongs to/equivalent a specific cluster then acuracy is 1␣
,→else 0

4 Exercise
1. What is the accuracy, precision, and recall of the model trained on Hand Written Digit Recog-
nition dataset?
2. Describe the dataset used in this lab exercise.
3. Give insights into the model trained for Hand Written Digit Recognition dataset.
4. Use the k-means algorithm in python to cluster the following 8 examples into 3 clusters:
A1=(2,10), A2=(2,5), A3=(8,4), A4=(5,8), A5=(7,5), A6=(6,4), A7=(1,2), A8=(4,9).
(a) Suppose that the centers of each cluster are A1, A4 and A7. Run the k-means algorithm
for 3 epochs only. At the end of this epoch show:
i. The new clusters (i.e. the examples belonging to each cluster)(mention the appro-
priate attribute used to identify the clusters in sklearn)
ii. The centers of the new clusters (mention the appropriate attribute used to identify
the cluster centers in sklearn)

Computer Engineering Department, Faculty of Technology, D. D. University Page 4 of 5

 Lab Manual: Machine Learning

5. Apply Elbow Method on Part A and Part B. How many clusters will you choose according
to elbow method?
6. Write the silhouette score for model trained in Part A and Part B using sklearn.

5 Reference
1. [Link]
2. [Link]
3. [Link]
[Link]

Computer Engineering Department, Faculty of Technology, D. D. University Page 5 of 5