First, we need to set up the viewer and server environment. To run the server, we need to start our environment by opening the System Console start the MONAI Label docker container:
docker run --gpus all --rm -ti --ipc=host --net=host -v /home/nvidia/datasets:/opt/monai/datasets projectmonai/monailabel:latest bash
Skip this step if you’ve already started the container for a previous app. There’s no need to start a new container if you already started one for another app. Stop previous server by pressing Ctrl+C and go to the next step.
After starting the docker container the next step is to download the Bundle and start the server using a sample dataset (MSD Task07 Pancreas dataset).
Download the Bundle:
monailabel apps --download --name monaibundle --output apps
To start the server, the MONAI Label Bundle app requires access to the MODEL Zoo. This means a GitHub authentication token has to be set. Please set the token as follows:
export MONAI_ZOO_AUTH_TOKEN=<Your github auth token>
If you don’t have a token, please check how to get one from here: https://docs.github.com/en/authentication/keeping-your-account-and-data-secure/creating-a-personal-access-token
Once we set the authentication token, we should be able to start the MONAI Label server using the following command:
monailabel start_server --app apps/monaibundle --studies datasets/Task07_Pancreas/imagesTr --conf models pancreas_ct_dints_segmentation_v0.3.5