In this notebook a toy example of pipeline for preparing variable length sequences is presented. They can be then used as input data e.g. to Recurrent Nerual Networks.
In the notebook three .csv files are loaded. First one contains stacked sequences of observations (each containing a set of features). The remaining two contain: length of following sequences and a target labels. Scheme explains how the lengths apply to stacked sequences:
