IssueTracker

Please describe your requested enhancement. Good feature requests will solve common problems or enable new use cases.

What you would like to accomplish:
Would like to be able to use Make AutoML Vision Object Detection Edge with Jetson Nano

How this might work:
Expand the GPU docker support to the device

If applicable, reasons why alternative solutions are not sufficient:
There is no known alternative soltuion at this point.

Other information (workarounds you have tried, documentation consulted, etc):

1) Followed the instructions here[1] but produced ".../cuda-10.0/compat/libcuda.so.410.48: no such file or directory" error when I run:

sudo docker run --runtime=nvidia --rm --name "${CONTAINER_NAME}" -v \
${YOUR_MODEL_PATH}:/tmp/mounted_model/0001 -p \
${PORT}:8501 -t ${GPU_DOCKER_GCS_PATH}

2) Built CUDA driver following [2]. The same error as above is produced.

[1] https://cloud.google.com/vision/automl/object-detection/docs/containers-gcs-tutorial
[2] https://github.com/NVIDIA/nvidia-docker/wiki/NVIDIA-Container-Runtime-on-Jetson