Transformations¶
The aeon.transformations
module contains classes for series
transformations. The module is organised into CollectionTransformers which transform a
collection of time series into a different representation and SeriesTransformers which
transform single time series.
All transformers in aeon can be listed using the aeon.registry .all_estimators utility, using estimator_types=”transformer” tag.
Collection transformers¶
Transformer base class for collections. |
|
Autocorrelation function transformer. |
|
Autoreggression coefficient feature transformer. |
|
Downsample the time dimension of a collection of time series. |
|
Discrete Wavelet Transform Transformer. |
|
HOG1D transform. |
|
Return the matrix profile and index profile for each time series of a dataset. |
|
Normaliser transformer for collections. |
|
Pad unequal length time series to equal, fixed length. |
|
Periodogram transformer. |
A transformer that turns time series collection into tabular data. |
|
|
Time series interpolator/re-sampler. |
|
StandardScaler for time series. |
|
Piecewise slope transformation. |
|
Truncate unequal length time series to a lower bounds. |
Channel selection¶
|
Performs channel selection using a single channel classifier or regressor. |
|
Elbow Class Pairwise (ECP) transformer to select a subset of channels. |
|
Elbow Class Sum (ECS) transformer to select a subset of channels/variables. |
|
Selects a random proportion of channels. |
Compose¶
|
Pipeline of collection transformers. |
Convolution based¶
|
RandOm Convolutional KErnel Transform (ROCKET). |
|
MINImally RandOm Convolutional KErnel Transform (MiniRocket). |
MINIROCKET (Multivariate, unequal length). |
|
|
Multi RandOm Convolutional KErnel Transform (MultiRocket). |
|
Hydra Transformer. |
Dictionary-based features¶
|
Symbolic Aggregate approXimation (SAX) transformer. |
|
Piecewise Aggregate Approximation Transformer (PAA). |
|
Symbolic Fourier Approximation (SFA) Transformer. |
|
Symbolic Fourier Approximation (SFA) Transformer. |
|
Bag-of-Receptive-Fields (BORF) Transformer. |
Feature based¶
|
Canonical Time-series Characteristics (Catch22). |
|
Transformer for extracting time series features via tsfresh.extract_features. |
|
Transformer for extracting time series features via tsfresh.extract_features. |
|
Seven-number summary transformer. |
Interval based¶
|
Random interval feature transformer. |
|
Supervised interval feature transformer. |
|
QUANT interval transform. |
Shapelet based¶
|
Random Shapelet Transform. |
Random Dilated Shapelet Transform (RDST) as described in [R1a26faa97573-1], [R1a26faa97573-2]. |
|
|
Scalable and Accurate Subsequence Transform (SAST). |
|
Random Scalable and Accurate Subsequence Transform (RSAST). |
Signature based¶
|
Transformation class from the signature method. |
Series transforms¶
|
Transformer base class for collections. |
|
Auto-correlation transformer. |
|
ClaSP (Classification Score Profile) Transformer. |
|
Filter a times series using Discrete Fourier Approximation (DFT). |
|
Distance based Outlier BasIs using Neighbors (DOBIN). |
|
Filter a times series using Gaussian filter. |
|
Calculate the matrix profile of a time series. |
|
Piecewise Linear Approximation (PLA) for time series transformation. |
|
Filter a times series using Savitzky-Golay (SG). |
|
Auto-correlation wrapper for statsmodels. |
|
Partial auto-correlation wrapper for statsmodels. |
|
Filter a times series using the Baxter-King filter. |
|
Box-Cox power transform. |
|
Distance based Outlier BasIs using Neighbors (DOBIN). |
|
Scaled logit transform or Log transform. |
|
Filter a times series using Recursive Median Sieve (SIV). |
|
Principal Components Analysis applied as transformer. |
|
Warping Path Transformer. |