Integrating Mqtt with neural network in Jetson TX 2

Hi guys,

I created a neural network weight of a sensor based dataset. I want to integrate this weight in MQTT py script to run real-time inference. How do I do this to give an output to be read by another script . Any assistance will be deeply appreciated.

Thanks

Hi,

Sorry that we don’t have too much experience on the MQTT.

There is no too much difference between desktop and TX2 in the python interface.
I think you can just try the sample of MQTT on the TX2 directly.

Thanks.

I’ve been able to Intergrate MQTT to Nvidia jetson which works well, that’s no issues, my problem is mqtt + neural network weight, how do I implement tht

Hi,

There is a sample implement different protocols (Kafka, AMQP, MQTT, Custom etc.) but only for the desktop user.
It is implemented as a gstreamer plugin. Maybe you can modify it for the Jeston.

https://developer.nvidia.com/deepstream-sdk-download
${DeepStream_Release}/sources/gst-plugins/gst-nvmsgbroker

Thanks.

Hi,
Where can I find the sample for AMQP, MQTT protocols for desktop user which you’ve mentioned?