Prasunet Internship Tasks

By: Nithin Kumar Reddy

This repository contains four machine learning projects completed as part of the Prasunet Internship. Each project tackles a different machine learning task, ranging from regression and clustering to image classification and food recognition.

---

Table of Contents

1. Task 1: House Price Prediction using Linear Regression
2. Task 2: Customer Segmentation using K-Means Clustering
3. Task 3: Cat vs. Dog Image Classification using SVM
4. Task 4: Food Recognition and Calorie Estimation

---

Task 1: House Price Prediction using Linear Regression

Problem Statement:

Develop a linear regression model to predict the prices of houses based on their square footage and the number of bedrooms and bathrooms.

Dataset:

• House Prices Dataset

Key Features:

• Simple regression model
• Predicts house prices based on house features
• Evaluates model using mean squared error (MSE)

---

Task 2: Customer Segmentation using K-Means Clustering

Problem Statement:

Create a K-Means clustering algorithm to group customers of a retail store based on their purchase history.

Dataset:

• Customer Segmentation Dataset

Key Features:

• K-Means clustering for customer segmentation
• Identifies groups of similar customers based on purchase behavior
• Visualizes clusters and customer data distribution

---

Task 3: Cat vs. Dog Image Classification using SVM

Problem Statement:

Implement a Support Vector Machine (SVM) to classify images of cats and dogs.

Dataset:

• Dogs vs. Cats Dataset

Key Features:

• SVM-based image classification
• Binary classification between two categories: cats and dogs
• Evaluates model accuracy on the test dataset

---

Task 4: Food Recognition and Calorie Estimation

Problem Statement:

Develop a model to accurately recognize food items from images and estimate their calorie content, enabling users to track their dietary intake.

Dataset:

• Food-101 Dataset

Key Features:

• Image classification of food items
• Calorie estimation for dietary tracking
• Recognizes various food items and provides calorie per gram estimates

Calories per gram for each food item:

• Apple Pie: ~2.5 calories per gram
• Baby Back Ribs: ~3.5 calories per gram
• Baklava: ~5 calories per gram
• Beef Carpaccio: ~2 calories per gram

---

How to Run the Projects

1. Clone the repository:

   git clone https://github.com/yourusername/prasunet-internship-tasks.git

2. Navigate to the desired task folder:

   cd task_1   # For Task 1
   cd task_2   # For Task 2
   cd task_3   # For Task 3
   cd task_4   # For Task 4

3. Install the required dependencies:

   pip install -r requirements.txt

4. Run the corresponding Python script for each task.

---

Acknowledgements

• Datasets from Kaggle are used for each project.
• These projects were completed as part of the Prasunet Internship Program.