Back to models
Classifier

MrSEQL

API referenceSource

MrSEQL = Multiple Representations Sequence Learning classification model.

Direct Interface to MrSEQLClassifier from mrseql. Note: mrseql itself is copyleft (GPL3). This interface is permissive license (BSD3).

MrSEQL is an efficient time series classifier utilizing symbolic representations of time series, using SAX and SFA features.

Quickstart

python
from sktime.classification.shapelet_based import MrSEQL

estimator = MrSEQL(seql_mode='fs', symrep='sax', custom_config=None)

Parameters(3)

seql_modestr, either ‘clf’ or ‘fs’ (default).
In the ‘clf’ mode, Mr-SEQL is an ensemble of SEQL models, while in the ‘fs’ mode Mr-SEQL, trains a logistic regression model with features extracted by SEQL from symbolic representations of time series.
symrepstr, or list or tuple of string, strings being ‘sax’ or ‘sfa’.
default = “sax”, i.e., only SAX features, no SFA features. The symbolic representations to be used to transform the input time series.
custom_configdict, optional, default=None
Customized parameters for the symbolic transformation. If defined, symrep will be ignored. (no documentation of this parameter is provided in the original mrseql code)

References

[1]

Thach Le Nguyen, Severin Gsponer, Iulia Ilie, Martin O’Reilly, Georgiana Ifrim. “Interpretable Time Series Classification Using Linear Models and Multi-resolution Multi-domain Symbolic Representations”, Data Mining and Knowledge Discovery, 2019.