Welcome to Jülich Data Challenges
About Us
Jülich Challenges is a platform for scientific data challenges initiated by Forschungszentrum Jülich. We host data challenges on the edge of scientific discovery promoting Jülich's unique scientific data and according research problems.
Jülich Challenges target ML researchers, or traditional image analysts, looking for interesting data to work on, that allow quick results for fast publications. As benefits for FZJ we expect to attract ML and AI specialists to our unique data sets, scientific questions and therewith to provide solutions for Jülich-specific research questions. This comes with an increased visibility in the fields of ML and AI - across domains and within the ML communities.
How to host a challenge on Jülich Data Challenge platform
-
Setup challenge configuration
We have created a sample challenge configuration that we recommend you to use to get started available at link. Clone this repository and follow the next steps. -
Edit challenge configuration
Openchallenge_config.yml
from the repository that you cloned in step-1. A challenge can be divided into many phases (or challenge phases) and a challenge phase can have the same or different start and end date than the challenge start and end date. Thechallenge_config.yml
file defines all the different settings of your challenge such as start date, end date, number of phases, and submission limits etc that allows to structure your challenge. Edit this file based on your requirement.You can refer to link for more details to configure the file based on your requirement.
-
Edit evaluation script
Next step is to edit the challenge evaluation scriptevaluation_script
that decides what metrics the submissions are going to be evaluated on for different phases. Theevaluation_script
evaluates the submission of participants and returns the scores which will populate the leaderboard. The logic for evaluating and judging a submission is customizable and varies from challenge to challenge.You can refer to link to edit the file based on your requirement.
-
Edit challenge HTML templates
Now, you just need to update the HTML templates in thetemplates/
directory of the bundle that you cloned. Each html files provides details such as challenge backgroud, evaluaton criteria, directions to download data, terms and condition, links to other useful resource, etc. to the participants. Yon can add text relevent to the your challenge. -
Upload configuration on Jülich Data Challenge platform
Finally run the./run.sh
script in the bundle. It will generate a challenge_config.zip file that contains all the details related to the challenge. Now, visit Jülich Data Challenge - Host challenge page and select/create a challenge host team. Then upload thechallenge_config.zip
.
How to participate in a Jülich Data Challenge
-
Visit Jülich Challenges
-
Sign up or Log in Sign Up and fill in your credentials or log in if you have already registered.
-
After signing up you would be on the dashboard page.
-
Select a challenge Then, go to challenges section and choose an active challenge.
-
Challenge Page After reading the challenge instructions on the challenge page, you can participate in the challenge.
-
Create Participant Team Create a participant team if there isn’t any or you can select from the existing ones. Click on ‘Participate’ tab after selecting a team.
How to submit your solution to a Jülich Data Challenge
The challenge you would like to contribute to gives you instructions how to download the respective data set. Several of them host their data directly on the challenges platform available at this link: link. Navigate there to download the files associated with the challenge. Each Jülich Data challenge comes with a zip files containing the datasets and information about the challenge and a Jupyter Notebook.
### Challenge Layout
data_public_leaderboard_phase.tar.gz #
notebook.ipynb # The notebook file.
The notebook.ipynb
contains the first steps to approach the problem that demonstrates a baseline solution. We hope that this will be a useful resource to you to explore the data and perform some data analysis and come up with a better solution. The notebook.ipynb
file also contains the code to generate the submision.csv
.
After generating the submision.csv
file, you can simply upload the csv file to the challenge frontend at link.