Instruction for Task 3 submissions

About evaluations in this task

In this task, the developed methods will be evaluated on the testing set by the organizers locally by running them within a docker provided by the challengers.

Docker instructions

You must provide a built image containing your method. Please refer to the Docker documentation to build your image.

During the evaluation, your docker will be executed by the organizers on the test dataset, and the output scores will be processed similarly as in task 2 to compute the C-index, using the following command line:

docker run -v /datalocation:/data -v /outputlocation:/output mydockerimage /bin/bash -c 
    "sh predict.sh /data/[PatientID]PET.nii.gz /data/[PatientID]CT.nii.gz /data/[PatientID]SegMask.nii.gz /output/output_score.csv”

Update: An example with Dockerfile and few sample scripts are provided here.

How to submit the submissions

1. Save your docker image:

docker save --output submission-v1.tar your_image_name

2. Upload the docker image as submission:

aicrowd submission create -c miccai-2021-hecktor -f submission-v1.tar

In case aicrowd-cli is not installed you can do so via pip install aicrowd-cli

You can download an archive containing the docker you can use to encapsulate your method here:

[download] https://discourse.aicrowd.com/uploads/short-url/4My8XdRJMhSDJynw5fELdFprV9h.zip
[mirror] https://filesender.renater.fr/?s=download&token=bf47950c-7bad-4e9c-aafe-dba761d77ddb

Please have a look at the README file for details

1 Like

I’m not able to get the example docker working.

process.sh and predict.sh are copied to /workspace in the image. But then in the readme, to run the container you mount a local directory to /workspace, and then process.sh and predict.sh are no longer accessible.

I think this could be fixed if instead of mounting to /workspace we mount to a different directory in the container to save the output. Like -v /outputlocation:/output

Then process.sh needs to be modified to change OUTPUT_file
OUTPUT_file="/output/$team.csv"

Thanks for the headsup we will correct this asap.

corrected docker file