AI Programming with Python Project

Introduction

Project code for Udacity's AI Programming with Python Nanodegree program. In this project, students first develop code for an image classifier built with PyTorch, then convert it into a command line application.

Goal

• Design a CNN using Pytorch to classify the flower data set.
• Train the CNN and achieve at least 70%
• Once the network is trained save the model
• Load the model and using an image from the test data set predict

Process

1. Train the network.
2. Save the checkpoint.
3. Load the checkpoint.
4. Using a filePredict the name.

Depedencies

1. Image dataset
2. Categories file

Files

1. train.py
2. predict.py
3. initialize.py
4. checkpoint.py

Reference

1. Pytorch documentation: Transfer learning
2. Udacity Transfer Learning Repo
3. Udacity : AI Python starting point Repo
4. Medium : Deep Learning With PyTorch by Josh Bernhard

Tools

1. Anaconda
2. VSCode