10
$\begingroup$

I have multiple separate time series and would like to train the same LSTM network on them. How to do in this situation? I can't just concatenate timeseries (along time), because I am afraid network will be confused by jumps at the points of concatenation.

How to overcome?

Improve this question
$\endgroup$
7
  • 2
    $\begingroup$ Can you clarify a little bit more? Do these time series have the same length? what do you want the RNN to do eventually etc. $\endgroup$ Commented May 30, 2018 at 10:23
  • $\begingroup$ All time series are of different lengths (coincidences are not excluded of course). The goal is to predict next one or more samples on the basis of previous ones. $\endgroup$ Commented May 30, 2018 at 11:20
  • $\begingroup$ Is there a reason for the length variation?, can't resample them to make them have the same number of samples? $\endgroup$ Commented May 30, 2018 at 11:55
  • $\begingroup$ @Amani they represent different periods of observations of some process; if I resample them, I will damage significant data $\endgroup$ Commented May 30, 2018 at 12:16
  • 1
    $\begingroup$ LSTM is invariant to 'speed'. Meaning, its strength is detecting patterns whether or not they have the same length. I thing it's worth trying at least. $\endgroup$ Commented May 30, 2018 at 14:02

2 Answers 2

Reset to default
5
$\begingroup$

just re-use

model.fit()

on the fresh datasets using the already trained model, simple as that :) !

(given that you do it in Keras)

Improve this answer
$\endgroup$
4
  • $\begingroup$ What will it do at the beginning? Make previous sample zero? $\endgroup$ Commented May 30, 2018 at 12:17
  • 1
    $\begingroup$ if it's part of the same code, the weights and the state will have the updated values, utilizing the learnt information from the previous training session. $\endgroup$ Commented May 30, 2018 at 12:22
  • $\begingroup$ But won't it have the same effect as if I concatenated series? $\endgroup$ Commented May 30, 2018 at 12:58
  • 3
    $\begingroup$ good idea, I guess it will have the similar "jump". Therefore consider model.reset_states() before feeding the new dataset and calling model.fit() :) $\endgroup$ Commented May 30, 2018 at 13:13
5
$\begingroup$

If I understand your question correctly, the reason you think cannot concatenate your time series into a one dataset is because of their different length. Depending on your problem, you can handle this issue in multiple manners in preprocessing. But the more common way is to use sequence padding. Preprocessing methods are natively implemented in keras: https://keras.io/preprocessing/sequence/.

Hope that answers your question.

EDIT 1:

The answer of @pcko1 anwser would work but'll force you into using a batch size of 1, which might give you a much higher convergence time.

Improve this answer
$\endgroup$
2
  • 1
    $\begingroup$ No, length is irrelevant, since I would concatenate series "horizontally", along time. $\endgroup$ Commented May 30, 2018 at 12:15
  • $\begingroup$ Could you reformulate your question to make sure that If I understand correctly, the different periodicities of your time series is your problem ? What I mean is that you have several time series data referring to the same problem but sampled differently ? Am I right ? $\endgroup$ Commented May 30, 2018 at 12:21

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.