Skip to content

roman-yang/Rot-CNN

Repository files navigation

Rot-CNN

Standard Convolutional Neural Networks (ConvNets) rely on data augmentation (particularly, rotating training images) to deal with rotation invariant pictures. We design a new convolution layer that is rotation invariant by nature. As a result, we don't need to rotate the training images for preprocessing.

Installation

Tested under Ubuntu with Python 3.10.

pip install -r requirements.txt

Usage

Execute main.py in each folder. The required datasets are automatically downloaded.

Here are several commands for you to use :

    --dataset 1     #choose dataset MNIST
    --dataset 2     #choose dataset FashionMNIST
    --train         #retrain a model, default : false, load trained model
    --augment       #do data augmentation, default : false

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages