Reproduced Papers

ReproducedPapers.org is an online repository of scientific papers reproductions with their codes. It aims to offer a platform both to share and access reproductions with both their codes and reproduction procedures (e.g. a blog post) in one place.

It is originally developed for CS4240 Deep Learning course of TU Delft by CV-Lab.

Installation

This application is written by using React and uses Firebase for backend and Algolia for search index. To locally run this application you need to follow below steps:

1. Install Node, Yarn and Firebase CLI. We are using Node version 14.13, Yarn version 1.22 and Firebase CLI version 8.12.
2. Clone this git repository to your computer by running git clone https://github.com/CVLab-TUDelft/reproduced-papers.git.
3. You need to create a project in each platform (one in Firebase and one in Algolia).
4. You also need to deploy firestore indexes and rules and storage rules to Firebase. To do this, run firebase deploy --only firestore:rules firestore:indexes storage:rules. The index creation may take sometime.
5. Copy the .env.example file and rename it as .env and write the needed configurations of the projects into the file.
6. Run yarn install to install the dependencies.
7. Finally, run yarn start to start the application.

Paper

We wrote a paper titled ReproducedPapers.org: an open online repository for teaching and structuring machine learning reproducibility over the value and the necessity of an online repository of reproductions. The paper was published at the RRPR 2020: Third ICPR Workshop on Reproducible Research in Pattern Recognition.

For the paper, we conducted two small anonymous surveys on two groups:

• students who recently added their reproduction to our repository,
• anybody identifying her/himself working in AI.

And here you can download the survey data: survey-data.zip.

Contribution

There are several ways of contributing:

1. Submitting your papers and reproductions to ReproducedPapers.org,
2. Sharing ReproducedPapers.org with your colleagues,
3. Improving this web application by adding features and fixing bugs.

Please don't hesitate to send pull request!