0
$\begingroup$

what is the best format to feed the input data, which are time series with varying density over time, to a deep learning network, while at any iteration we want to feed a batch of data including a historical background?

Is it better to consider a constant size of data records or a constant time window including variable data record size? Or is there a better way?

$\endgroup$
0
$\begingroup$

Not sure if this is will help, but in a similar problem I did the following:

  1. Normalized the data
  2. Created lags (as many as the desired inputs + 1 more for the output)
  3. Split the dataset into training and testing
  4. Created the model by feeding it with the inputs form the training test
  5. Got the prediction results from the test set
  6. Un-normalized the data
  7. Ran the metrics (RMSE, MAPE etc) to test the accuracy
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.