I am working on a time series forecasting problem and I am in the process of choosing the optimum network structure. Currently I have a 200 cell LSTM layer fully connected to 100 neurons in an adjacent layer, which in turn to a 24 neuron output layer.

My prediction problem is time series forecast of GHI values in a 24 hour window using previous 24 hours as input.

I would like to know whether making the second layer as LSTM cells will make any difference?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.