Dog Identification Application

This project is a part of the Deep Learning Nanodegree at Udacity

-- Project Status: Completed

Project Introduction

Building a pipeline to process real-world, user-supplied images. Given an image of a dog, the algorithm will identify an estimate of the canine’s breed. If supplied an image of a human, the code will identify the resembling dog breed.

Methods Used

• Deep Learning
• Convolutional Neural Networks
• Transfer Learning

Technologies

• Python 3
• anaconda or miniconda
• PyTorch
• Numpy
• OpenCV
• tqdm
• PIL
• torchvision
• matplotlib
• jupyter notebook

Project Description

The purpose of the project is to develop an algorithm that could be used as part of a mobile or web app that will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. OpenCV's implementation of Haar feature-based cascade classifiers is used to detect human faces in images. VGG-16 model is used to detect dogs in images. I used Transfer Learning to create dog breed classifier that got 0.55 test loss and 83% test accuracy after 30 epochs of training. (for more details see this documentation)

Getting Started

1. Clone this repo (for help see this tutorial).
2. Install the above technologies
3. Create a new conda environment >> conda create --name deep-learning python=3
4. Enter the environment: (Mac/Linux) >> source activate deep-learning, (Windows) >> activate deep-learning
5. Run the following to open up the notebook server >> jupyter notebook