Dog Breed Identification

Welcome to the Dog Breed Identification App, a deep learning-powered web application built with TensorFlow, Keras, and Flask. Designed for image classification, this project aims to accurately detect and identify 120 dog breeds from raw images using a Convolutional Neural Network (CNN) trained on the official Kaggle dataset.

🚀 Project Overview

The Dog Breed Identifier is a machine learning model trained via supervised learning on a labeled dataset of over 10,000 images spanning 120 distinct dog breeds. It extracts unique breed-specific features through a convolutional architecture and predicts the breed of dogs in unseen images. The final result is served via an intuitive web interface powered by Flask.

✨ Features

120 Dog Breeds Supported – Covers a wide range of common and rare dog breeds
Real Image Input – Upload an image of any dog to get breed predictions
Deep CNN Model – Built using TensorFlow and Keras with high accuracy
Interactive Flask Web App – Easy-to-use interface for predictions
Kaggle Dataset Integration – Based on the official Kaggle competition
Model Performance Reporting – Shows breed confidence score

🛠️ Tech Stack

Component	Details
Language	Python
Framework	Flask
ML Libraries	TensorFlow, Keras, NumPy, Pandas
Model Type	Convolutional Neural Network (CNN)
Dataset	Kaggle Dog Breed Identification
Deployment	Localhost via `flask run`

📷 UI Screenshots

🐶 Homepage

📸 Prediction Result

📦 How to Run

✅ Requirements

Python 3.8+
Virtual environment tools (venv/pipenv)
All required packages in requirements.txt

🔧 Setup Instructions

Clone the Repository

git clone https://github.com/abdulrehmandev/dog-breed-classifier.git
cd dog-breed-classifier

Create & Activate Virtual Environment

python -m venv venv
source venv/bin/activate   # On Windows use: venv\Scripts\activate

Install Dependencies
```
pip install -r requirements.txt
```
Run the App
```
flask run
```

💡 Make sure to download the dataset separately from Kaggle and place it in the appropriate directory before training or prediction.

📝 You can also take model training Notebook from Kaggle

🧠 Educational Purpose

This project was built to reinforce deep learning fundamentals such as:

Image classification using CNNs
Data preprocessing and augmentation
Model evaluation and fine-tuning
Serving ML models through web applications

Perfect for ML beginners interested in combining computer vision and deployment.

Discoverability

Dog Breed Classifier, TensorFlow CNN Projects, Image Recognition Flask App, Deep Learning Dog Image Classifier, Kaggle Dog Breed Challenge, Breed Identification AI, Python Image Classifier, Computer Vision Projects with Flask.

🤝 Contribution

Contributions, feature requests, or ideas are welcome! Fork the repo, create a branch, and submit a pull request. Let’s improve this together.

📜 License

This project is licensed under the MIT License. See the LICENSE file for details.

👨‍💻 Author

Abdul Rehman

abdulrehmandev/dog-breed-identifier