LSTM-Neural-Network-for-Time-Series-Prediction
LSTM-Neural-Network-for-Time-Series-Prediction copied to clipboard
jaungiers
future predictions
How do you predict future data let's say for a year? i see how you take observed values, and then compare the results of the model to the actual to determine how accurate you are, but i dont see how you can easily say, predict for next year
You can predict a data sequence for a year, but it will be slightly inaccurate... ;)
In general, load your model and use the function "predict_point_by_point". Iterate in a for loop for a given window_size (a year for instance... depends what timestemp did you used during trainig, so you will have to recalculate it). Store and shift the current_predicted_window for a new prediction and re-run the prediction. At the end your model will predict a state based on its own prediction (therefore i would not recommend such an approach) :)
