evaezekwem / flower-image-classifier-cnn

This repo contains data set and code to train a convolutional neural network to correctly identify 102 different species of flowers.

Flower-Image-Classifier-CNN

This repo contains data set and code to train a convolutional neural network to correctly identify 102 different species of flowers.

To Use Anaconda (Recommended)

• Install Anaconda from here

To install on cpu (not the best option)

• conda install pytorch-cpu torchvision-cpu -c pytorch

To install on gpu

• Go to the Pytorch page

Launch Jupyter from command prompt/terminal

• From command prompt jupyter notebook
• Navigate to the project folder and open "Image Classifier.ipynb"

To Use Pip (Not recommended)

To install dependencies

pip install requirements.txt

To train the model

python image_classifier.py

To Simply Load the Model

• Navigate to the cell block in the notebook and run the cell then test the model.