DogBreedClassifier

This project features a deep learning model designed to classify various dog breeds from images. Created as part of my personal journey through the Fastai deep learning course, it aims to enhance my skills with the Fastai API. The model is trained on the Stanford Dogs Dataset and can accurately identify 120 dog breeds based on visual features. This repository includes code for training, evaluating, and deploying the model, along with links to the dataset of dog images.

Features:

• Pre-trained deep learning models for dog breed classification.
• Support for transfer learning to improve model accuracy.
• Easy-to-use scripts for training and deploying the model on HuggingFace.

Folder Structure

• Production: Contains the code and materials necessary for deploying the model on HuggingFace.
• Train: Includes the code and dataset required to train the model. Open dog-breeds-classifier.ipynb to train the model or use the pre-trained model dog_breeds_clf_model.pkl.

The HuggingFace web app for this project is available here.

The dataset can be found on kaggle here or from the original data source here.