RADAR.time_series.preprocessing
Submodules
RADAR.time_series.preprocessing.preprocessing_ts module
- class RADAR.time_series.preprocessing.preprocessing_ts.FilterPreprocessing(filter_func=numpy.mean, kernel_size=3, **kwargs)[source]
Bases:
BasePreprocessing- transform(X)[source]
Apply filtering using a moving average kernel.
Input: - X: numpy.ndarray or pandas.Series, representing a one-dimensional
time series. If a DataFrame is provided, each column will be processed independently.
Output: - numpy.ndarray or pandas.Series with the filtered series, using
a moving average filter with the specified kernel size.
- class RADAR.time_series.preprocessing.preprocessing_ts.InterpolationPreprocessing(method='linear', **kwargs)[source]
Bases:
BasePreprocessing- transform(X)[source]
Apply interpolation to fill missing values.
Input: - X: pandas.DataFrame or pandas.Series, where rows are time steps
and columns may contain missing values that need to be interpolated.
Output: - pandas.DataFrame or pandas.Series with interpolated values, using
the specified interpolation method (e.g., ‘linear’, ‘polynomial’).
- class RADAR.time_series.preprocessing.preprocessing_ts.MinMaxScalerPreprocessing(**kwargs)[source]
Bases:
BasePreprocessing
- class RADAR.time_series.preprocessing.preprocessing_ts.NormalizerPreprocessing(**kwargs)[source]
Bases:
BasePreprocessing
- class RADAR.time_series.preprocessing.preprocessing_ts.OneHotEncoderPreprocessing(columns=None, **kwargs)[source]
Bases:
BasePreprocessing
- class RADAR.time_series.preprocessing.preprocessing_ts.RobustScalerPreprocessing(**kwargs)[source]
Bases:
BasePreprocessing
- class RADAR.time_series.preprocessing.preprocessing_ts.RollingMeanPreprocessing(window=3, **kwargs)[source]
Bases:
BasePreprocessing