tsfresh already calculates a comprehensive number of features. If you are interested in which features are calculated, just go to our
tsfresh.feature_extraction.feature_calculators
module. You will find the documentation of every calculated feature there.
tsfresh enforces a strict naming of the created features, which you have to follow whenever you create new feature calculators. This is due to the tsfresh.feature_extraction.FeatureExtractionSettings.from_columns method which needs to deduce the following information from the feature name
- the time series that was used to calculate the feature
- the feature calculator method that was used to derive the feature
- all parameters that have been used to calculate the feature (optional)
Hence, to enable the tsfresh.feature_extraction.FeatureExtractionSettings.from_columns to deduce all the necessary conditions, the features will be named in the following format
{time_series_name}__{feature_name}__{parameter name 1}_{parameter value 1}__[..]__{parameter name k}_{parameter value k}
(Here we assumed that {feature_name} has k parameters).
So for example the following feature name
temperature_1__quantile__q_0.6
is the value of the feature tsfresh.feature_extraction.feature_calculators.quantile for the time series `temperature_1 and a parameter value of q=0.6`. On the other hand, the feature named
Pressure 5__cwt_coefficients__widths(2, 5, 10, 20)__coeff_14__w_5
denotes the value of the feature tsfresh.feature_extraction.feature_calculators.cwt_coefficients for the time series `Pressure 5 under parameter values of widths=(2, 5, 10, 20), coeff=14 and w=5`.