Using RNNs to do sequence to sequence modeling for power consumption values in kW/hr. Comparing predicted value (PV) from RNN to exact power consumption value (EV) to find the deflection between PV and EV.
power_forecast.py : initial training.
testing.py : checking accuracy.
more_training.py : further training done using pre-trained weight file and different optimizers.
training_data.npz : Consists training data with keywords train_X and train_Y.
log_file : Keeps track of all the results and updates done to the model.