I am trying to build a neural network that has an input of n pairs of integer values (where n is random) and a corresponding output of binary array with n length.
Input will be a set of integer value coordinates [(x1, y1), (x2, y2), (x3, y3), …, (x50, y50), ...] where each instance can be of various length like [(x1, y1), (x2, y2), (x3, y3), …, (x52, y52)] or [(x1, y1), (x2, y2), (x3, y3), …, (x101, y101)], etc.
Output is a set of binary arrays with each instance having the same length as the corresponding input.
May i know if anyone has any recommendations on what neural network would fit this use case?
Thanks alot!
You can try RNN structures, like LSTM and GRU. You can treat each of the tuple (x_i, y_i) as an input for a specific time-step. RNN structure accepts input with unfixed time-length, can by setting 'return_sequence = True', the corresponding output will have the same length as the input.
Hope this helps.
Dear Kian Pee ,
Look the link, maybe useful.
https://stackoverflow.com/questions/35675440/arbitrary-length-of-input-output-for-recurrent-neural-network-lstm
Regards,
Shafagat
- Badges
- Science topic