Transformer
ClaSPTransformer
ClaSP (Classification Score Profile) Transformer.
Implementation of the Classification Score Profile of a time series. ClaSP hierarchically splits a TS into two parts, where each split point is determined by training a binary TS classifier for each possible split point and selecting the one with highest accuracy, i.e., the one that is best at identifying subsequences to be from either of the partitions.
Quickstart
python
from sktime.transformations.clasp import ClaSPTransformer
estimator = ClaSPTransformer(window_length=10, scoring_metric='ROC_AUC', exclusion_radius=0.05)Tags
Capabilities
- Unequal-length series
- Multivariate: Not supported
- Inverse transform: Not supported
- Missing values: Not supported
- Removes missing values: Not supported
- Equalizes series length: Not supported
Properties
- Input typescitype:transform-input
- Series
- Output typescitype:transform-output
- Series
- Label typescitype:transform-labels
- None
- Fit is emptyfit_is_empty
- Yes
- Keeps the time indextransform-returns-same-time-index
- No
- Requires Xrequires_X
- Yes
- Requires yrequires_y
- No
- X and y need the same indexX-y-must-have-same-index
- No
Parameters(3)
- window_lengthint, default = 10
- size of window for sliding.
- scoring_metricstring, default = ROC_AUC
- the scoring metric to use in ClaSP - choose from ROC_AUC or F1
- exclusion_radiusint
- Exclusion Radius for change points to be non-trivial matches
Examples
>>> from sktime.transformations.clasp import ClaSPTransformer
>>> from sktime.detection.clasp import find_dominant_window_sizes
>>> from sktime.datasets import load_electric_devices_segmentation
>>> X, true_period_size, true_cps = load_electric_devices_segmentation ()
>>> dominant_period_size = find_dominant_window_sizes (X)
>>> clasp = ClaSPTransformer (window_length = dominant_period_size)
>>> clasp. fit (X)
>>> profile = clasp. transform (X)