Extract Cell Array from Varying Sequence Length Dataset

Hi,

does anyone here know how the Cell Array in the JapanesVowels Dateset is extracted from the Dataset. I have a similar Problem and want to classify a Dataset of varying length sequences but i dont know how to extract the data into such a cell Array

You can find the Cell Arry by using this Command in Matlab Online.

[XTrain,YTrain] = japaneseVowelsTrainData;

It uses the following Dataset

https://archive.ics.uci.edu/ml/datasets/Japanese+Vowels

Answer 1

[Gaurav Garg]

Hi,

The Japanese vowels training set contains 270 sequences. Each sequence has 12 feature dimensions and a variable sequence length. Responses are a categorical vector of labels "1","2",...,"9", which correspond to nine speakers.