Monai Usage

From UFRC
Revision as of 16:41, 2 August 2023 by Dimi (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Back to Monai

MONAI core

NGC container usage:

ml purge
ml ngc-monai/<version>
python <your__python_script>

MONAI label

NGC container usage:

1. Server - to start the server as a slurm job:

  • Load modules:
  ml purge
  ml ngc-monailabel/<version>
  • Copy to your directory the file:
  /apps/nvidia/containers/monai/start_monai_server_readonly.sh
  • Copy e.g. these directories to a place you own (may vary by use case):
  cp -r /apps/nvidia/containers/monai/apps/deepedit <my_place>
  cp -r /apps/nvidia/containers/monai/datasets/Task09_Spleen <my_place>
  • Modify the start_monai_server_readonly.sh line to read:
  singularity exec -B /apps/nvidia/containers/monai /apps/nvidia/containers/monai/monailabel/ monailabel start_server --app <my_place>/deepedit --studies <my_place>/Task09_Spleen/imagesTr
  or, for a newer version e.g. (the apps/... and datasets/... directories may be different)
  singularity exec -B /apps/nvidia/containers/monai /apps/nvidia/containers/monai/monailabel.0.6.0/0.6.0 monailabel start_server --app <my_place>/deepedit --studies <my_place>/Task09_Spleen/imagesTr
  • Start server as a batch job:
 sbatch start_monai_server_readonly.sh
  • Note server address from job output: used in next step

2. 3DSlicer client

  • Start Open On Demand (OOD) session
  • Start Console in hwgui with 1 GPU: gpu:geforce:1
  • In console: load & start Slicer
 ml qt/5.15.4 slicer/4.13.0
 vglrun -d :0.$CUDA_VISIBLE_DEVICES Slicer
  • In Slicer GUI:
    • Select module: Active Learning -> MONAILabel
    • Fill in server address, e.g.: http://c1007a-s17:8000/
    • Click on refresh button next to server address
    • Load Next Sample
  • You are good to go! Enjoy!