I'm quite new in ANNs. I intend to use ANNs for predicting spike points in time series right before they happen. I've already used LSTM for another scenario, and I know that they can be used in similar situations as well.

Can anyone give me a piece of advice or some suitable resources that might be used as a beginning point? It would be much appreciated if it uses DeepLearning4J for implementation.

  • 1
    $\begingroup$ Since you mentioned that you have already used LSTMs I can directly guide you to using it for time Series prediction. For detailed description you can check the following link: <towardsdatascience.com/…> This is a very useful article for you to grasp the idea behind time Series Prediction. I suggest you go through this to get an idea and then try implementing this in dl4j. $\endgroup$
    – SUMAN
    Aug 8, 2019 at 8:28
  • $\begingroup$ What kind of time series data is it? Do you happen to have covariates? What is the sampling interval of the data? How big is the data -- i mean number of data points? Multivariate..? $\endgroup$
    – naive
    Aug 10, 2019 at 10:29


You must log in to answer this question.

Browse other questions tagged .