Transformer
ShapeletTransform
Shapelet Transform.
Original journal publication: @article{hills2014classification,
title={Classification of time series by shapelet transformation}, author={Hills, Jon and Lines, Jason and Baranauskas, Edgaras and Mapp, James and Bagnall, Anthony}, journal={Data Mining and Knowledge Discovery}, volume={28}, number={4}, pages={851–881}, year={2014}, publisher={Springer}
}
Schnellstart
python
from sktime.transformations.shapelet_transform import ShapeletTransform
estimator = ShapeletTransform(min_shapelet_length=3, max_shapelet_length=inf, max_shapelets_to_store_per_class=200, random_state=None, verbose=0, remove_self_similar=True)Parameter(12)
- min_shapelet_lengthint, lower bound on candidate
- shapelet lengths (default = 3)
- max_shapelet_lengthint, upper bound on candidate
- shapelet lengths (default = inf or series length)
- max_shapelets_to_store_per_classint, upper bound on number of
- shapelets to retain from each distinct class (default = 200)
- random_stateRandomState, int, or none: to
- control random state objects for deterministic results (default = None)
- verboseint, level of output printed to
- the console (for information only) (default = 0)
- remove_self_similarboolean, remove overlapping
- “self-similar” shapelets from the final transform (default = True)