LSTMs for Time Series Forecasting in Matlab 2017b

I have a series of queries:- 1] How can I perform time series forecasting i.e. given a time series of lets say 'x' time steps and predicting the next 'y' time steps (consecutively). Can I use the regression layer after the last layer or will I have to convert my time series problem as a classification problem?

2] What is the significance of padding?

3]How do I prepare a dataset from a time series to be put into use for training the LSTM network?

4]Within the documentation for LSTM layers, under the sub-topic: 'list parameters' what Output Mode (sequence/ last) should I select and when?

[nora mahmoud]

did you reach to solution , if you reached please help me , as i have the same problems..

Answer 1

[Abolfazl Nejatian]

here is my code,

https://www.mathworks.com/matlabcentral/fileexchange/69506-time-series-prediction

this piece of code predicts time series data by use of deep learning and shallow learning algorithm.

best wishes

Abolfazl Nejatian