Building Real-time Dermatology Classification with NVIDIA Clara AGX

Originally published at: https://developer.nvidia.com/blog/building-real-time-dermatology-classification-with-nvidia-clara-agx/

The most commonly diagnosed cancer in the US today is skin cancer. There are three main variants: melanoma, basal cell carcinoma (BCC), and squamous cell carcinoma (SCC). Though melanoma only accounts for roughly 1% of all skin cancers, it is the most fatal, metastasizing rapidly without early detection and treatment. This makes early detection critical,…

For those wishing to run this container on the Clara AGX/Holoscan dGPU, here is a workaround:

First - run the Dermatology container with the added option: -v /media/m2:/m2

We’ll need to copy the following folders to the m2 drive:

cp models/ /m2/ -r

cp source/ /m2/ -r

Now pull the AGX PyTorch Container:

docker pull nvcr.io/nvidia/clara-agx/agx-pytorch:21.05-1.7-py3

export DISPLAY=:0
xhost +
mkdir /home/nvidia/results

sudo docker run --gpus all -it --rm -e DISPLAY=$DISPLAY --device /dev/video0:/dev/video0 -v /media/m2:/m2 -v /tmp/.X11-unix:/tmp/.X11-unix nvcr.io/nvidia/clara-agx/agx-pytorch:21.05-1.7-py3

apt-get update
apt-get install python3-setuptools ffmpeg libsm6 libxext6 libhdf5-dev opencv-python -y
pip3 install --upgrade pip
pip3 install efficientnet-pytorch

cp /m2/models/ . -r
cp /m2/source/ . -r
cd source/
python3 demo.py