Time series classification#
The sktime.classification
module contains algorithms and composition tools for time series classification.
All classifiers in sktime
can be listed using the sktime.registry.all_estimators
utility,
using estimator_types="classifier"
, optionally filtered by tags.
Valid tags can be listed using sktime.registry.all_tags
.
A full table with tag based search is also available on the Estimator Search Page (select “classifier” in the “Estimator type” dropdown).
Composition#
|
Pipeline of transformers and a classifier. |
|
Applies estimators to columns of an array or pandas DataFrame. |
|
Pipeline of transformers and a classifier. |
|
MultiplexClassifier for selecting among different models. |
Model selection and tuning#
|
Exhaustive search over specified parameter values for an estimator. |
Ensembles#
|
Bagging ensemble of time series classifiers. |
Time Series Forest Classifier as described in [R880353f3a8bd-1]. |
|
|
Weighted ensemble of classifiers with fittable ensemble weight. |
Deep learning#
|
Time Convolutional Neural Network (CNN), as described in [Rd06fa525366f-1]. |
|
Contextual Time-series Neural Classifier (CNTC), as described in [1]. |
|
Fully Connected Neural Network (FCN), as described in [R4ddccc18e565-1]. |
|
Implementation of LSTMFCNClassifier from Karim et al (2019) [1]. |
|
Gated Recurrent Unit (GRU) for time series classification. |
|
GRU-FCN for time series classification. |
|
InceptionTime Deep Learning Classifier. |
|
Multi-Scale Attention Convolutional Neural Classifier, as described in [R18bfe0ce893e-1]. |
|
Multi Layer Perceptron Network (MLP), as described in [R2c17e28b4d16-1]. |
|
Multi Channel Deep Convolutional Neural Classifier, as described in [Re4d070570206-1]. |
|
Multivariate Time Series Transformer for Classification, as described in [Rd7b568e6af84-1]. |
|
Residual Neural Network as described in [1]. |
|
Simple recurrent neural network. |
|
Time series attentional prototype network (TapNet), as described in [Rdbe61d137f31-1]. |
Dictionary-based#
|
Bag-of-SFA Symbols in Vector Space, from pyts. |
|
Ensemble of Bag of Symbolic Fourier Approximation Symbols (BOSS). |
|
Contractable Bag of Symbolic Fourier Approximation Symbols (cBOSS). |
|
Single bag of Symbolic Fourier Approximation Symbols (IndividualBOSS). |
|
Single TDE classifier, an extension of the Bag of SFA Symbols (BOSS) model. |
|
MUSE (MUltivariate Symbolic Extension). |
|
Temporal Dictionary Ensemble (TDE). |
|
Word Extraction for Time Series Classification (WEASEL). |
Distance-based#
|
The Elastic Ensemble (EE). |
K-nearest neighbours Time Series Classifier. |
|
K-nearest neighbors time series classifier, from |
|
K-nearest neighbors Time Series Classifier, from tslearn. |
|
|
Proximity Forest classifier. |
|
Proximity Stump class. |
|
Proximity Tree class. |
|
ShapeDTW classifier. |
Dummy#
|
DummyClassifier makes predictions that ignore the input features. |
Early classification#
Probability Threshold Early Classifier. |
|
|
Two-tier Early and Accurate Series Classifier (TEASER). |
Feature-based#
|
Canonical Time-series Characteristics (catch22) classifier. |
|
Fresh Pipeline with RotatIoN forest Classifier. |
|
Martrix Profile (MP) classifier. |
|
Random Interval Classifier. |
|
Classification module using signature-based features. |
|
Summary statistic classifier. |
|
Time Series Feature Extraction based on Scalable Hypothesis Tests classifier. |
Hybrid#
|
Hierarchical Vote Collective of Transformation-based Ensembles (HIVE-COTE) V1. |
|
Hierarchical Vote Collective of Transformation-based Ensembles (HIVE-COTE) V2. |
Interval-based#
|
Canonical Interval Forest Classifier (CIF). |
|
Diverse Representation Canonical Interval Forest Classifier (DrCIF). |
Random Interval Spectral Ensemble (RISE). |
|
|
Supervised Time Series Forest (STSF). |
|
Time series forest classifier. |
Kernel-based#
|
Support Vector Classifier, for time series kernels. |
|
Time Series Suppoer Vector Classifier, from tslearn. |
|
Arsenal ensemble. |
|
Classifier wrapped for the Rocket transformer using RidgeClassifierCV. |
Shapelet-based#
|
A shapelet transform classifier (STC). |
|
Learning Shapelets algorithm, from pyts. |
Learning Time Series Shapelets Classifier, from tslearn. |
|
|
MrSEQL = Multiple Representations Sequence Learning classification model. |
|
MrSQM = Multiple Representations Sequence Miner. |
sklearn classifiers#
This section contains classifiers which are not time series classifiers but
simple tabular classifiers in sklearn
compatible API.
They are used internally in time series classifiers, but can also be used directly in a tabular setting.
|
Continuous interval tree (CIT) vector classifier (aka Time Series Tree). |
|
A rotation forest (RotF) vector classifier. |
Base#
Abstract base class for time series classifiers. |
Abstract base class for deep learning time series classifiers. |
Abstract base class for early time series classifiers. |