Use the keras timeseriesgenerator for sktime? Is the timestamp necessary?

  dataloader, generator, keras, python, sktime

I have a dataframe with some (35 columns) Variables and millions (rows) of timesteps. I would like to cut the data with the timeseriesgenerator (https://www.tensorflow.org/api_docs/python/tf/keras/preprocessing/sequence/TimeseriesGenerator). I have done it before for some neural nets. I would like to stick to the generator, because I need to skip some samples, but how many isn’t clear at the moment.

Is it possible to use the generator also to provide the examples for sktime?
I have a classification problem. The data is provided in a long mode. The generator does not provide timestamps. Or is there any workaround?
I would like to use sktime, but failed to get the data in shape. The data contains a time variable, but there was a problem while recording. So there are big gaps, but this was just a saving issue.

If necessary I would add some sample data in an hour or so.

Source: Python Questions

LEAVE A COMMENT