I built a feedforward neural network, his goal is to classify a target value, the train has to be defined for n-days.
The input is constituted by 12 neurons (since I have 12 different data each of which for each day has around 25000 values [25000 x 1]). The network, for every single day, manages 12 * 25000 input values. (The data are all numeric)
Since I have to train for about 741 different days (This implies I have to perform a train with 741 * 12 * 25000 data), at the moment I'm able to train only for a single day, save the model and reload it to train for the subsequence day.
There is a way to avoid this save - load model procedure and train for all the 741 days at once? Should I have to use a particular data generator?
Can anyone give me some suggestions?
Thank you very much.
