rsantana-isg / Adversarial-Crafting-FashionMNIST

Adversarial examples crafting for Fashion MNIST

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Crafting adversarial examples for the fashion MNIST dataset

A program that generates adversarial examples for the Fashion MNIST dataset to fool a Convolutional Neural Network for Machine Learning and Neural Networks (MLNN) course.

Author

Eritz Yerga Gutierrez - eritzyg

Dataset

The dataset used for this project is the Fashion MNIST dataset. The dataset must be downloaded and placed in a specific folder (explained in the notebook in detail).

How to run

To run this project you need Python and Jupyter Notebook. Also you will need to install all the python libraries specified in the notebook.

Documentation

Documentation is available in the releases section with the corresponding release.

Results

Notice about pre-trained CNN

To facilitate the usage of the program, a pre-trained CNN is included with the notebook. The training has been done with 30 epoches and a batch size of 100.

About

Adversarial examples crafting for Fashion MNIST

License:MIT License


Languages

Language:Jupyter Notebook 100.0%