I have a multivariate time series (3D) dataset of the shape, (samples, sequences, features). For instance, (patients, time_sequences_in_hour, features), whereby each sample is a distinct group of time sequences. For each sample, there is a binary or multiclass target.
I read through questions such as Feature selection for time series data, but the question covers time series forecasting which libraries like TsFresh can handle. Besides, the feature selection libraries in Python that I have surveyed so far are on tabular (2D) data, i.e: ReliefF.
Is there a feature selection library in Python that works well for multivariate time series classification dataset as I mentioned above?
