AI AND MACHINE LEARNING LAB

Structure Page
No.
2.0 Introduction 18
2.1 Objectives 18
2.2 General Guidelines 18
2.3 Python Libraries for AI and Machine Learning 19
2.4 Working with Python 21
2.5 Running Python Programs 24
2.6 List of Lab Assignments – Session wise 25
2.0 INTRODUCTION
This is the lab course, wherein you will have the hands on experience towards various
Artificial intelligence & Machine Learning algorithms,You have studied in MCS-
224(Artificial intelligence & Machine Learning) course of this programme.

This course explores the implementation of various Artificial intelligence & Machine
Learning algorithms, using Python programming under Google Colabs or ANACONDA
or any other Python programming environment. A list of programming problems is also
provided at the end of each session. Please go through the general guidelines and the
program documentation guidelines carefully.

The pre-requisites of the course includes the understanding of Python programming, you
may refer to the concepts of Python programming given in the course MCS-
201(Programming in C & Python) and also try to attempt the lab exercises given in the
section-2 of MCSL-205(C and Python Lab.).

In order to have better understanding you are advised to firstly get the conceptual clarity of
the various algorithms presented in the MCS-224(Artificial intelligence & Machine
Learning) course of this programme, and then implement your understanding in Python
programming, to solve the problems given in the various sessions of this course.

2.1 OBJECTIVES
After completing this lab course you will be able to:

 Understand the Algorithms for Artificial Intelligence (AI), from the

perspective of programming
 Understand the Algorithms for Machine Learning(ML), from the
perspective of programming
 Understand the real power Python programming

2.2 GENERAL GUIDELINES

 You should attempt all problems/assignments given in the list session
wise.

 You may seek assistance in doing the lab exercises from the concerned
lab instructor. Since the assignments have credits, the lab instructor is
obviously not expected to tell you how to solve these, but you may ask
questions concerning the algorithmic implementation of the concepts
of AI and ML in Python programming language or any technical
problem.
 For each program you should add comments above each function in
the code, including the main function. This should also include a
description of the function written, the purpose of the function,
meaning of the argument used in the function and the meaning of the
return value (if any).

 These descriptions should be placed in the comment block

immediately above the relevant function source code.

 The comment block above the main function should describe the
purpose of the program. Proper comments are to be provide where and
when necessary in the programming.

 The program should be interactive, general and properly documented

with real Input/ Output data.

 If two or more submissions from different students appear to be of the

same origin (i.e. are variants of essentially the same program), none of
them will be counted. You are strongly advised not to copy somebody
else's work.

 It is your responsibility to create a separate directory to store all the

programs, so that nobody else can read or copy.

 Observation book and Lab record are compulsory

 The list of the programs (list of programs given at the end, session-
wise) is available to you in this lab manual. For each session, you must
come prepare with the algorithms and the programs written in the
Observation Book. You should utilize the lab hours for executing the
programs, testing for various desired outputs and enhancements of the
programs.

 As soon as you have finished a lab exercise, contact one of the lab
instructor / in-charge in order to get the exercise evaluated and also get
the signature from him/her on the Observation book.

 Completed lab assignments should be submitted in the form of a Lab

Record in which you have to write the algorithm, program code along
with comments and output for various inputs given.

 The total no. of lab sessions (3 hours each) are 20 and the list of
assignments is provided session-wise. It is important to observe the
deadline given for each assignment.

2.3 PYTHON LIBRARIES FOR AI AND

MACHINE LEARNING

Python programming is widely used in Artificial Intelligence, Machine

Learning, Neural Networks and many other advanced fields of Computer
Science. Ideally, It is designed for rapid prototyping of complex
applications. Python has interfaces with various Operating system calls and
libraries, which are extensible to C, C++ or Java. Many large companies
like NASA, Google, YouTube, Bit Torrent, etc. uses the Python
programming language for the execution of their valuable projects.

We learned from the comparison of C and Python, given in MCS – 201

(Programming in C & Python), that Python is a high-level, general-
purpose, interpreted high level programming language. It is dynamically
typed and garbage-collected, and supports multiple programming
paradigms like structured (particularly, procedural,) object-oriented,
and functional programming, and due to its comprehensive standard library
Python is often described as a "batteries included" language

Python is the most popular programming language, and it marked its

presence by contributing actively to every emerging field in computer
science. It has vast set of libraries for almost all fields such as Machine
Learning (Numpy, Pandas, Matplotlib), Artificial intelligence (Pytorch,
TensorFlow), and Game development (Pygame,Pyglet), and many more.

We are already aware that the development of different type of applications

require different types of packages, libraries, modules, frameworks etc.
Now you might be confused that what is the difference between these
terms, are they same or different. Let’s Clear your confusion first and then
we will briefly discuss about the functionality of different packages, used
for the development of various Python applications.

Python: Framework Vs Library Vs Package Vs Module

Framework is a collection of libraries. This is the architecture of the

program.

Library is a collection of packages. (We may understand that, Python

library or framework is a pre-written program that is ready to use on
common coding tasks.)

Package is a collection of modules. It must contains an __init__.py file as a

flag so that the python interpreter processes it as such. The __init__.py
could be an empty file without causing issues. A package, in essence, is
like a directory holding subpackages and modules. While we can create our
own packages, we can also use one from the Python Package Index (PyPI)
to use for our projects.To import a package, we type import
Game.Sound.load Or We can also import it giving it an alias:import
Game.Sound.load as load game.

Module is a file which contains various Python functions and global

variables. It is simply just .py extension file which has python executable
code. We put similar code together in one module. This helps us
modularize our code, and make it much easier to deal with. And not only
that, a module grants us reusability. With a module, we don’t need to write
the same code again for a new project that we take up.
let’s see how Module and Package differ:

1) A module is a file containing Python code. A package, however, is like

a directory that holds sub-packages and modules.

2) A package must hold the file __init__.py. This does not apply to
modules.

3) To import everything from a module, we use the wildcard *. But this

does not work with packages.

We will learn more about them, as we proceed in this course, don’t worry.
To start with we will discuss about various Frameworks and Libraries first,
you will be learning about their usage and also the usage of methods and
packages, later.

An introduction towards the functionality of some of the libraries, which

are frequently used in the areas of Artificial Intelligence and Machine
Learning, is given below :

 TensorFlow: TensorFlow is an end-to-end python machine learning

library for performing high-end numerical computations.
TensorFlow can handle deep neural networks for image recognition,
handwritten digit classification, recurrent neural networks, NLP
(Natural Language Processing)

 Keras : is a leading open-source Python library used for

development of neural networks and machine learning projects. It
simplifies the process of designing and development of neural
networks for the beginners of machine learning. Keras also deals
with convolution neural networks(CNN) and Recurrent Neural
Networks (RNN), highly required in the field of Deep Learning. It
includes algorithms for normalization, optimizion, and activation
layers. Instead of being an end-to-end Python machine learning
library, Keras acts as a user-friendly, extensible interface that
enhances modularity & total expressiveness.

 Theano : is a library for scientific computation, it allows you to

define, optimize as well as evaluate the complex mathematical
expressions, which deals with multidimensional arrays. The
repetitive computation of a tricky mathematical expression is the
basis of several ML and AI applications. Theano quickly performs
the data-intensive calculation, the rate of execution is almost
hundred times faster than when executing on our CPU alone. It
aims to reduce the development and execution time of ML apps,
particularity in deep learning algorithms. Only one drawback of
 Theano in front of TensorFlow is that its syntax is quite hard for the
beginners.

 Scikit-learn : it is a prominent open-source library for machine

learning through Python, it includes a wide range of algorithms like
DBSCAN, gradient boosting, random forests, vector machines, and
k-means etc., which are generally used to implement various
concepts of Machine Learning i.e. Classification, Clustering,
Regression etc. It can interoperate with numeric and scientific
libraries of Python like NumPy and SciPy. Scikit-learn supports
both supervised as well as unsupervised ML.

 PyTorch : it is a production-ready library for machine-learning,

supported with excellent examples, applications and use cases, also
supported by a strong community.It supports GPU and CPU
computations, thus provides performance optimization and
scalablity in research as well as production. The two high-end
features of the PyTorch are Deep neural networks and Tensor
computation, which are boosted because of the machine learning
compiler “Glow”, specially designed to improve the performance of
deep learning frameworks.

 NumPy : also known as Numerical Python, a library to perform

scientific computations. Almost all Python machine-learning tools
like Matplotlib, SciPy, Scikit-learn, etc rely on this library to a
reasonable extent. It comes with functions for dealing with complex
mathematical operations like linear algebra, Fourier transformation,
random number and features that work with matrices and n-arrays
in Python. It is widely used in handling sound waves, images, and
other binary functions.

 SciPy : It is library, that works for all type od scientific

programming projects. Its main functionality is built upon NumPy,
and it includes modules for linear algebra, Integration, optimization,
and statistics too. SciPy is supported with extensive documentation,
which makes its working really easy.

 Pandas : it is an open-source library of python that offers a wide

range of tools for data analysis & manipulation, with Pandas, the
data from a broad range of sources like CSV, SQL databases, JSON
files, and Excel, can be read. The data analysis & manipulation is a
pre-requisite for most of the machine learning projects, where a
significant amount of time is spent to analyse the trends and
patterns hidden in the datasets. Using Pandas one can manage
 complex data operations with just one or two commands, it comes
with several inbuilt methods for data handling, and it also serves as
the starting point to create more focused and powerful data tools.

 Matplotlib : this library is specifically meant for Data Science, it

helps to generate data visualization like 2D diagrams and graphs
like histograms, scatter plots, and even graphs for non-cartesian
coordinates.It is equipped with the object oriented API to embed
plots into the applications, directly. It is because of this library, the
Python is able to compete with scientific tools like MATLAB or
Mathematica. It is quite compatible with the popular plotting
libraries, but developers are required to write more code than usual,
while using this library for generating advanced visualizations.

 Seaborn : it is an unparalleled visualization library, based on

Matplotlib’s foundations. Seaborn offers high-level dataset based
interface to make amazing statistical graphics. With Seaborn, it is
simple to create certain types of plots like time series, heat maps,
and violin plots. The functionalities of Seaborn go beyond Pandas
and matplotlib with the features to perform statistical estimation at
the time of combining data across observations, plotting and
visualizing the suitability of statistical models to strengthen dataset
patterns.

 Scrapy : It is the most widely used library of python, for the

purpose of data sciences, specifically data mining. It is used to build
crawling programs i.e spider bots, which are used to retrieve
structured data (example URLs or Contact Information) from the
web. Developers use it for gathering data from APIs.

 Pygame : is a python programming language library for making

multimedia applications like video games, animations etc. It
includes libraries related to computer graphics and sound, which are
designed to be used with the Python programming language.
Pygame is suitable to create client-side applications that can be
potentially wrapped in a standalone executable.

 Django : it a web framework build using Python,it encourages rapid

development of web applications.Developers can focus on writing
app only, It’s primary goal is to ease the creation of complex,
database-driven websites. Django emphasizes on code
 reusability and pluggability of components, thus it supports less
code, low coupling, and rapid development of websites.

 MicroPython : is a software implementation of Python 3, written in

C. Infact, MicroPython is a Python compiler that runs on the
hardware of micro-controller, it includes modules to provide access
to the low-level hardware. One of the renowned embedded
application is Raspberry Pi which uses Python for its computing. It
can be used as a computer or like a simple embedded board to
perform high-level computations, you can use it for IoT or Mobile
applications, thus may produce your smart gadgets.

 BeeWare – is a Python GUI and mobile development framework, to

develop native Python Mobile Apps. “BeeWare” project, provides a
set of tools and an abstraction layer, which can be used to write
native-looking mobile and desktop applications using Python.

 Kivy – is an opensource Python Framework used as a Cross-

platform Python GUI for Mobile App development, it allows you to
write pure-Python graphical applications that run on both i.e. the
main desktop platforms (Windows, Linux, macOS and Raspberry
Pi) and also on on iOS & Android.Both Kivy and BeeWare are
worth considering. So far as maturity goes, Kivy seems to be the
more mature platform right now

In this course you are given exposure to implement the various algorithms
of Artificial intelligence & Machine Learning (You have studied in MCS-
224) by using various libraries of Python as programming language.

2.4 WORKING WITH PYTHON

Now, we are having a bit of clarity about the Python as a programming
language, from the past sections we learned about the various libraries and
frameworks of Python. Now, we need to work on the IDEs (Integrated
Development Environments) of Python, there are many IDEs like Jupyter
Notebook, Spyder, VS Code, R Programming etc., all are collectively
available in Anaconda (Anaconda is a free and open-source distribution of
the Python and R programming languages for scientific computing (data
science, machine learning applications, large-scale data processing,
predictive analytics, etc.)), or you may also go for the cloud versions Like
Google Colab Notebook, where you need not to have high configuration
hardwares, only internet is required and through your gmail account you
can work on Python using Jupyter notebook.
We will discuss in brief, some of the ways to work with Python, you may
choose any or try all and other options too.
1) Just browse for https://www.python.org and perform following steps :

a) Download the latest version of Python for the operating system

installed on your computer, from
https://www.python.org/downloads/

b) Now Run this exe file and install the Python, just click next and go
ahead, till the setup installation is finished

c) Finally you are having an interface for Python programming

There is a variety of IDEs for python, Like Jupyter Notebook, Spyder, VS
Code etc, and all are available at Anaconda (a free and open-source
distribution of the Python and R programming languages for scientific
computing (data science, machine learning applications, large-scale data
processing, predictive analytics, etc.)). To start with Anaconda Just perform
following steps.
2) Just browse https://www.anaconda.com/ and perform following steps:

a) Click the Download button on the webpage of

https://www.anaconda.com/
b) the distribution section https://www.anaconda.com/distribution/
will open,

c) click the download option given on this page

https://www.anaconda.com/distribution/
d) It is recommended to download 64bit version of Python 3.x

e) Now, just run the setup, and click next-next, till the installation is
completed.

f) Finally, you will find Anaconda Navigators shortcut on your

desktop, click on it and you can start working with any IDE be it
Jupyter Notebook, Spyder, VS Code etc., even you can work with
R-Programming.

3) Many a times the learners may not be equipped with the systems
having latest hardware configuarations, as desired for the
installation of Python, or their might be compatibility issues with
operating syatem or may be due to any reason you are not able to
install and start your work with Python. Under such circumstances
the solution is Google Colab Notebook
(https://colab.research.google.com /notebooks/ welcome.ipynb), use
this and just login with your gmail account and start your work on
Jupyter Notebook, as simple as that.
2.5 RUNNING PYTHON PROGRAMS
Just click file option and select new workbook, and new Jupyter notebook
will open in Google Colab, now you may start your work.

To begin with just copy paste the code given below, in Google Colab :

(Note : It is a program in Python to understand how Python's Scikit-Learn

library can be used to implement the KNN algorithm.)

# Import necessary modules

from sklearn.neighbors import KNeighborsClassifier
from sklearn.model_selection import train_test_split
from sklearn.datasets import load_iris
# Loading data
irisData = load_iris()
# Create feature and target arrays
X = irisData.data
y = irisData.target
# Split into training and test set
X_train, X_test, y_train, y_test = train_test_split(
X, y, test_size = 0.2, random_state=42)
knn = KNeighborsClassifier(n_neighbors=7)
knn.fit(X_train, y_train)
# Predict on dataset which model has not seen before
print(knn.predict(X_test))
Execute it by simply pressing the play button, you may see that the output
comes just beneath, as shown below:

Note :

 You may refer to the python codes given in Unit no. 16 of MCS-224
 You may refer to various links given at section 2.7