Flowers-Image-Classifier

• The project represents Flowers Image Classifier that can classify 102 different types of flowers from their images using PyTorch and transfer learning.
• The dataset used in this project can be found here.
• This project was done as a part of Udacity's AI Programming with Python Nanodegree.

Parts of the project

• The project can be separated into two parts:

  1. Jupyter Notebook:

     • We will implement the model in a jupyter notebook to experiment different models .ipynb.
       • Flowers Image Classifier.ipynb

  2. Python Command Line Application:

     • We will convert the code to python application that run from the command line to be embedded in other applications .py.
       • train.py
       • predict.py
       • model_functions.py
       • utility_functions.py

  Note: cat_to_name.json contains a dictionary mapping the integer encoded categories to the actual names of the flowers

Dependencies

• PyTorch 0.4.0
• Torchvision 0.2.1
• Numpy
• MatPlotLib
• PIL
• json

Note : Using the values of hyperparameters and learning rate written in this project with VGG16 model will result in 92% accuracy.