chengyuhsieh / explanation_robustness Goto Github PK

Python 4.69% HTML 83.07% Jupyter Notebook 12.24%

explanation_robustness's Introduction

The main file is "mnist_example.ipynb". Please use Jupyter notebook to open the file.
We also provide an HTML file "mnist_example.html" containing the executed results of the notebook. The file could be open in standard web browsers.

The code is written in Python 3.
- Install python dependencies:
1. `pip install -r requirements.txt`

Reference:
1. https://github.com/PAIR-code/saliency
2. https://github.com/marcoancona/DeepExplain
3. https://github.com/ruthcfong/pytorch-explain-black-box
4. https://github.com/chihkuanyeh/saliency_evaluation

explanation_robustness's People

Contributors

Stargazers

Watchers

explanation_robustness's Issues

Potential issue regarding corrupted pretrained weight files

Hi, thanks for the great work!

I am trying to reproduce your example notebook (mnist_example.ipynb), but I notice several files are 2240KB, and these files seem to be corrupted. I am wondering if you could kindly re-upload or point me to anywhere to have these files?

I would really appreciate your help and love to learn from your work.

The files are:
mnist/MNIST/processed/test.pt
mnist/MNIST/processed/training.pt
mnist/MNIST/raw/train-images-idx3-ubyte.gz (This can be downloaded from http://yann.lecun.com/exdb/mnist/)

Many thanks in advance!

Recommend Projects