/DogBreedClassification

Primary LanguageJupyter Notebook

Dog Breed Classifier

Neural network that classifies the top 120 dog breeds

Table of Contents
  1. About The Project
  2. Installation
  3. Contributing
  4. Usage
  5. Contact
  6. Acknowledgements

About the Project

Dog breed classifier built with the ResNet architecture and ImageNet transfer learning. This model was trained on the Stanford dog breed dataset from Kaggle with the fastai library for pytorch. After fitting the frozen ResNet50 model with ImageNet weights for 8 cycles, an accuracy of 0.904276 was achieved.

Built With

Installation

  • pip
    git clone https://github.com/danielchang2002/DogBreedClassification.git

    pip install -r requirements.txt

Contributing

  1. Fork it (https://github.com/yourname/yourproject/fork)

  2. Create your feature branch (git checkout -b feature/fooBar)

  3. Commit your changes (git commit -am 'Add some fooBar')

  4. Push to the branch (git push origin feature/fooBar)

  5. Create a new Pull Request

    View Demo · Report Bug · Request Feature

Usage

screenshot of demo

Deployed live at: https://dog-neural-net.herokuapp.com/

Contact

Daniel Chang - @twitter_handle - danielchang2002@gmail.com

Project Link: https://github.com/danielchang2002/DogBreedClassification

Acknowledgements