Invalid network, machine learning, conactenating layers, unconnected input

I'm trying to build a machine learning network to train against image data that has additional features. For example, I may have a picture of a plane and also know that the plane is 5 km away with an angular elevation of 20 degrees. My networks starts by working on the image data and then I want to introduce the feature data. I've been trying to follow the example given at https://www.mathworks.com/help/deeplearning/ug/train-network-on-image-and-feature-data.html without much luck. I currently have one known problem and a second unknown problem. The first problem is that I get an error saying that I have an invalid network and that each layer input must be connected to the output of another layer, but the graph of the layers looks good as I understand it.

The second problem is that I have no idea how to introduce the features data into the process. The example uses data stores which I'm not using. Somehow the features associated with each image have to get passed to the trainNetwork function. When using just the images, trainNetwork accepts the image stack as the input but now I have two sets of data to pass to the function. Any ideas about the invalid network and how to pass both sets of data to trainNetwork?

Answer 1

[JG]

The example in the link is for 2022a. I had been running Matlab 2021a. The errors were becuase the 2022a implementaion doesn't work with 2021a.