Every datapoint in my dataset consists of 3 time series. The data in the time series is discretized into equal time-bins but the 3 time series were measured for varying length. Time series 1 has 10 bins, series 2 5 bins and series 3 only 1 single bin. So an exmaple datapoint looks like this

1, 4, 1, 7, 3, 7, 3, 7, 3, 1
9, 6, 4, 7, 1

I would like to run two 1D convolutional layers across all of these timeseries and then combine them with a fully connected layer. Now how would I go about this since the data consists of three entries with varying input sizes? I do think that it makes sense to share the weights of the conv layer between series 1 and 2 as they are fundamentally similar.

Note that I simplified the data for this example and the real data is much longer than 10, 5, 1.



Your Answer

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

Browse other questions tagged or ask your own question.