🐶 Dog Breed Classification (120 Breeds) | Deep Learning End-to-End Project

This repository contains a complete end-to-end deep learning pipeline for multi-class image classification of 120 dog breeds using TensorFlow, Keras, and Python.
The project demonstrates the full workflow — from data preprocessing and augmentation to model training, evaluation, and deployment — following modern best practices in computer vision and machine learning.

---

🚀 Project Overview

The goal of this project is to build a multi-class image classifier capable of identifying a dog's breed from an image out of 120 possible categories.
The dataset is based on the Stanford Dogs Dataset, which includes over 10,000 images representing 120 breeds.

---

🧠 Key Features

• 🧩 120-Class Classification (multi-class softmax)
• 🖼️ Deep CNN Models: Transfer Learning with EfficientNet, ResNet50, and InceptionV3
• ⚙️ End-to-End Pipeline:
  • Data Collection & Preprocessing
  • Augmentation & Normalization
  • Model Training & Fine-tuning
  • Evaluation & Visualization
  • Model Export & Deployment