I'm trying to create a flow on my Raspberry PI 4 with TensorFlow nodes, but I can't make them work.
As you can see in the screenshot below, I'm testing 2 different TensorFlow nodes:
- the one named "tfjs object detection" (node-red-contrib-tfjs-object-detection), that is stuck in "loading model" status. I didn't pass any model file to him, because I don't have .json files to pass, so I let the node look for a model from a public endpoint via web;
- the other one named "tf coco ssd" (node-red-contrib-tfjs-coco-ssd), that should automatically work with coco-ssd model, but it doesn't work too. As you can see from the debug, it shows me a SyntaxError soon after the Deploy.
I linked the outputs of the 2 nodes to a function node, that should filter the detected objects, and a msg.payload debug node, that I use to understand if the 2 orange nodes work.
How can I fix one of them? I didn't pass any model, but I'd like to understand how to pass a model.json file to one of them. Thanks in advance for the help