Something that I personally use is Google Trends. This is a very useful tool for verifying the interest of a broad public on some subject. Results can even be refined to include region and/or time span.
For instance, here you can see a comparison for the interest in Tensorflow, Keras and Pytorch over the past 12 months:
There is no label for such bounding boxes, they are simply "ignored" during training. You can assign any value for their "labels", then multiplying what ever loss these boxes generated with 0. If there is no loss, there is no gradient from these boxes.
You can do that by defining a count_boxes vector with binary values. Object and ...
Instead of using a token embedding you can use a linear layer. For an input of (10, 5, 4) - (sequence length, batch size, features) you can create a linear layer:
self.embedding_layer = nn.Linear(4, d_model)
Where d_model is the dimension of the input to the transformer.
PositionalEncoding is still needed so as to have a representation of time in the inputs....