I don't know exactly how docker container will be run and the exact input, output & ground truth path

I don't know exactly how docker container will be run and the exact input, output & ground truth path  

  By: mtoan65 on Aug. 17, 2024, 1:22 p.m.

Hello, I encountered an issue during the Preliminary Development Phase CT Submission despite successful local testing of my repository and Docker container. Upon reviewing the logs, I noticed that the input and output folders were empty. This leads me to believe there might be a discrepancy between my local setup and the submission environment. Specifically, I'm unclear about:

  • The runtime configuration or pipeline for the Docker container in the submission environment.
  • The expected directory structure and exact paths for input, output and ground truth files.
 Last edited by: mtoan65 on Aug. 17, 2024, 1:22 p.m., edited 1 time in total.

Re: I don't know exactly how docker container will be run and the exact input, output & ground truth path  

  By: YSang on Aug. 17, 2024, 6:22 p.m.

Hello. The input and output paths are mounted through docker volume, and should be managed in exactly the same way as the example algorithm (set by inference.py and test_run.sh).

Participants do not have access to the ground truths. They are only to be used in the evaluation, not the algorithm itself.

Re: I don't know exactly how docker container will be run and the exact input, output & ground truth path  

  By: mtoan65 on Aug. 19, 2024, 10:23 a.m.

Do we need to define the docker volume in docker file ? Sorry I'm just a docker newbie :(