If my data structure is is a single numerical vector what kind of network structure would I use?
For example, if I wanted to classify between 4 different car types and the data is the cars speed at each min for 4 hours.
In my mind, it seems like it might depend on if you believe that there is a relationship between one data point and its surrounding data points and lead you to use an RNN. Is this true - is there an advantage to an RNN or CNN in these cases? Would it even matter?
Sorry if this is a redundant question, or if I’m not using the term “Rank 1 Tensor” correctly.