Jetson Nano 2gb giving runtime errors while training

I am actually quite new to this platform.I get the following error which is presented in the screenshot attached while creating a model and training it using a Raspberry pi v2 camera in jupyter notebook and the live output wouldn’t get displayed in the final widget .I just get a still frame in the output widget.I have tried increasing the swap memory to 8 gb and even tested it in headless mode but the same error persists.I have tested the camera using the hello camera notebook previously and it works fine.
Any help would be appreciated.
Thank you in advance

Hi,

Could you restart the docker with --memory=500M --memory-swap=8G configure.

For example:
1. Create 8GB swap
2. Replace the below command in the docker_dli_run.sh and then restart.

sudo docker run --runtime nvidia -it --rm --network host \
    --volume ~/nvdli-data:/nvdli-nano/data \
    --device /dev/video0 \
    --memory=500M \
    --memory-swap=8G \
    nvcr.io/nvidia/dli/dli-nano-ai:v2.0.1-r32.6.1"

Thanks.

Hi,
I have followed your steps and have made the changes yet it gives out the same error and couldn’t get a live feed from my camera in the widgets at the end just as it was before.
And i use a raspberry pi csi v2 camera. I have appended my docker dli script as below:

sudo docker run --runtime nvidia -it --rm --network host \
    --volume ~/nvdli-data:/nvdli-nano/data \
    --volume /tmp/argus_socket:/tmp/argus_socket \
    --device /dev/video0 \
    --memory=500M --memory-swap=8G \
    nvcr.io/nvidia/dli/dli-nano-ai:v2.0.1-r32.6.1

Any help would be appreciated.

UPDATE --Ok ,so I tried it in headless mode again seems to work most of the times but still couldn’t get the live feed in the widget when using the GUI.