I have a time dataset of 43680 columns and 2205 rows. I am trying to use tsfresh to extract relevant features through default_fc_parameters = EfficientFCParameters(). I have set the n_jobs to 4. I am having trouble with the chunk size. Does increasing the chunk size increase the processing speed ?
X = extract_features(df, column_id="id", column_sort="T", default_fc_parameters = EfficientFCParameters(), n_jobs = 4, chunksize = 200)
