wwalterr / flower

Flower classification using VGG and Resnet

Vida

Flower classification using Convolutional Neural Networks through PyTorch

The result of this project was 97% accuracy using Resnet 18 and 92% using VGG 11, to see the result open the folders resnet18 and vgg11.

Dependencies

Install the Pyenv and Pyenv Virtualenv.

Install Anaconda image:

pyenv install anaconda3-2018.12

Create a virtual environment:

pyenv virtualenv anaconda3-2018.12 vida

Activate the virtual environment:

pyenv activate vida

Install the dependencies:

conda install --yes --file requirements.txt

List installed modules:

conda list

Update conda:

conda update -n base -c defaults conda

Run

Start Jupyter Lab:

jupyter lab

Credits

Github.com/Sphinxs

Robots.ox.ac.uk/~vgg/data/flowers/102