National University of technology

Name: Muhammad Safi Ullah Adam

ID: F20604031
Batch: CEN-20
Subject: ML & AI LAB

Submitted to: Engr. Abdul Qadeer

 Experiment 7

Multi-class Classification and Neural Networks

Objectives
The objective of this lab is to
 Learn and understand the concept of Logistic Regression with Regularization.
 Implement Logistic and regularization Algorithm in Python.

Software:

 Python Idle 3.6 or 3.9

 Jupyter Notebook

Theory:
> In this exercise, you will implement one-vs-all logistic regression and neural networks to recognize
hand-written digits.

## 1. Multi-class Classification

> For this exercise, you will use logistic regression and neural networks to recognize handwritten digits
(from 0 to 9). Automated handwritten digit recognition is widely used today - from recognizing zip codes
(postal codes) on mail envelopes to recognizing amounts written on bank checks. This exercise will show
you how the methods you’ve learned can be used for this classification task.
In the first part of the exercise, you will extend your previous implementation of logistic regression and
apply it to one-vs-all classification.
### 1.1 Dataset

> You are given a data set in ex3data1.mat that contains 5000 training examples of handwritten digits,
where each training example is a 20 pixel by 20 pixel grayscale image of the digit.

Steps:
Conclusion

Utilizing neural networks for multi-class classification tasks offers robustness and adaptability in tackling
complex categorization problems. With their ability to discern complex patterns and create intricate
representations, neural networks excel in classifying input data into various categories accurately. This
method improves the sophistication and adaptability of classification models, making them exceptionally
appropriate for a wide array of applications. These include image and speech recognition, natural
language processing, medical diagnosis, and many other fields that require nuanced differentiation
between multiple categories. By capturing subtle distinctions in data, neural networks facilitate deeper
insights and more nuanced predictions, positioning them as an indispensable tool in the ever-evolving
landscape of machine learning and artificial intelligence.

Harnessing the power of neural networks in the realm of multi-class classification affords unparalleled
robustness and a dynamic approach to complex sorting tasks. Neural networks' inherent capacity to
interpret convoluted patterns and engineer detailed representations equips them with the finesse needed
for assigning a range of classes to incoming data.