Description

DeepLabCut is a toolbox for markerless pose estimation of animals performing various tasks.

Software Category: bio

For detailed information, visit the DeepLabCut
website.

Available Versions

To find the available versions and learn how to load them, run:

module spider deeplabcut

The output of the command shows the available DeepLabCut
module versions.

For detailed information about a particular DeepLabCut
module, including how to load the module, run the module spider command with the module’s full version label. For example:

module spider deeplabcut/2.2.1.1-anipose

Dockerfile

We cannot use the official Docker image on the UVA HPC system because:

• the CUDA version is incompatible with our NVIDIA driver version (as of August 2021);
• at runtime it tries to download pretrained models inside the container, which is not possible via Apptainer.

For further details please visit here.

Usage

Python script

Please submit jobs to the GPU partition. A Slurm script template is provided below.

#!/bin/bash
#SBATCH -A mygroup      # your allocation account
#SBATCH -p gpu          # partition
#SBATCH --gres=gpu:1    # number of GPUs
#SBATCH -N 1            # number of nodes
#SBATCH -c 1            # number of cores
#SBATCH -t 10:00:00     # time

module purge
module load apptainer deeplabcut

apptainer run --nv $CONTAINERDIR/deeplabcut-2.2.1.1-anipose.sif myscript.py

GUI

Please request a Desktop session on the GPU partition via our Open OnDemand portal. Open a terminal and load the module. Then execute:

module load apptainer deeplabcut
apptainer run --nv $CONTAINERDIR/deeplabcut-2.2.1.1-anipose.sif -m deeplabcut