I am trying to make an NN for meteorological prediction, for which I have input data of only one meteorological station. I have target data of more than one location.
I have to train the NN in such a way that I gove it two input values (e.g.current temperature, pressure) and want to obtain temperature output at more than 50 locations and more than one time steps.
e.g. the input :
(25, 101.32)
should give output:
temperature after 2 hrs = 25, 26, 28, 29, 27.5
temperature after 4 hrs = 24, 23, 26, 26.5, 27
In which pattern to arrange the input and target data, and how to change the number of NN output nodes to obtain these results?