Should size of training vector and Input vector be matched in DTW algorithm?

irllabs / ml-lib

A machine learning library for Max and Pure Data

Other

274 stars 40 forks source link

Should size of training vector and Input vector be matched in DTW algorithm? #168

Closed jss8882 closed 4 years ago

jss8882 commented 4 years ago

I wanna use DTW algorithm to match a pattern which has different size. However, only i can see is error message that "error: ml.dtw: invalid input length, expected 8, got 80" how can i solve this problem?

NiccoloGranieri commented 4 years ago

Hi @jangsuJang, thanks for using ml.lib. From what I understand you are training [ml.dtw] with a feature vector of size 8, and after training the model you are feeding it a vector of size 80. To my knowledge DTW can be used to classify any type of N-dimensional temporal signals but the size of the signal must remain consistent between the training data and the mapping data.

jss8882 commented 4 years ago

Thank you for answering

NiccoloGranieri commented 4 years ago

No problem. Let me know if you need any more help.