ANFIS - different length vectors as training data

Hey everyone,

I have a question regarding the ANFIS training data. What I am trying to do is to train the ANFIS to classify different EMGs.

I have the data (compressed with wavelet transform), some coefficients and the output/class that the specific EMG belongs to.

The Problem is, that the recorded EMGs vary in length! Some are 2 sec, others 3 sec and so on.

Has anyone got an idea on how I can train the ANFIS with this data?

Cheers

Answer 1

[Riados]

What I thought of is padding the shorter vectors with zeroes, or reducing the bigger vectors to the length of the smaller ones by taking a mean to get them all to the same length.