Skip to content

COMP6258-Reproducibility-Challenge/reproducibility-challenge-benchmarking-attribution-methods

Repository files navigation

Prerequisites

Make sure to download this repo with submodules

The idsds repo is a submodule in this repository. In order to download any repo with submodules, run:

git clone --recurse-submodules https://github.com/your/repo.git

Download ImageNet-mini

import kagglehub
path = kagglehub.dataset_download("ifigotin/imagenetmini-1000")
print("Path to dataset files:", path)

Save the dataset to datasets/imagenet-mini/

Download CIFAR-100

CIFAR-100 download page

Download fine-tuned models

Download links are in idsds/README.md.

Create an environment

Create conda environment with:

conda env create -f idsds/conda_environment_export.yml --name reproducing_idsds

To activate this environment, use

conda activate reproducing_idsds

To deactivate an active environment, use

conda deactivate

Running the experiments

For each of our experiments, the files are available in a subdirectory of experiments/ folder. To run the experiments, follow the instructions in the README.md file for chosen experiment.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors