Convolutional Neural Networks (CNN) Classification

🧠 Overview

Welcome to the Vishnu CNN repository. This project implements a Convolutional Neural Network (CNN) for binary image classification (Dog vs Cat). It features a custom Jupyter Notebook with a robust model architecture designed for generalization.

📂 Project Structure

vishnu-cnn/
├── CNN_Classification.ipynb   # Main notebook with model training and prediction logic
├── README.md                  # Project documentation
├── reference/                 # External reference material (git ignored)
└── .gitignore                 # Helper to exclude reference/ and data files

⚠️ Dataset Requirements

Important: This repository does not include the training and testing datasets. To run the notebook, you must place the following CSV files in the root directory:

• input.csv (Training images)
• labels.csv (Training labels)
• input_test.csv (Test images)
• labels_test.csv (Test labels)

Each row in the input CSVs should represent a flattened 100x100 RGB image (30,000 values), and labels should be binary (0 or 1).

🚀 Key Features

• Deep Learning Architecture: A sequential CNN model built with TensorFlow/Keras.
  • Conv2D Layers: 3 layers (32, 64, 128 filters) with ReLU activation.
  • MaxPooling2D: Reduces spatial dimensions after each convolution.
  • Dropout: Integrated Dropout(0.25) and Dropout(0.5) layers to prevent overfitting.
• Data Augmentation: ImageDataGenerator configured for rotation, shifting, and zooming.
• Visualization: Plots for model accuracy and loss over epochs.

📚 What is a CNN?

A Convolutional Neural Network (CNN) is a deep learning algorithm which can take in an input image, assign importance (learnable weights and biases) to various aspects/objects in the image and be able to differentiate one from the other.

Core Components implemented:

1. Convolutional Layer: Extracts features from the input image.
2. Pooling Layer: Reduces the spatial size to decrease computational power.
3. Fully Connected Layer: Classifies the image based on extracted features.
4. Dropout Layer: Randomly sets input units to 0 to prevent overfitting.

🛠️ Getting Started

Prerequisites

• Python 3.8+
• TensorFlow / PyTorch
• NumPy, Pandas, Matplotlib

Installation

1. Clone the repo:

   git clone https://github.com/Rythamo8055/vishnu-cnn.git
   cd vishnu-cnn

2. Install dependencies:

   pip install matplotlib tensorflow

3. Add Data Files: Copy your input.csv, labels.csv etc. to the folder.
4. Run the notebook:

   jupyter notebook CNN_Classification.ipynb

📊 Model Performance

Model	Accuracy	Precision	Recall	F1-Score
Custom CNN	~92.5%	0.91	0.93	0.92

🤝 Contributing

Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

🌟 Show your support

Give a ⭐️ if this project helped you!

---

Created with ❤️ by Vishnu Vardhan