Which algorithm is best for fast time series classification?

For a balance of speed and high accuracy, ROCKET variants like MiniRocketClassifier are generally recommended for fast prototyping.

Can I use Aeon with standard scikit-learn tools?

Yes, Aeon is built to be scikit-learn compatible, meaning you can use its estimators and transformers inside standard sklearn Pipelines and GridSearchCV.

What kind of data format does the Aeon skill expect?

Aeon typically expects time series data in a 3D NumPy array format with the shape (n_samples, n_channels, n_timepoints).

Does Aeon support deep learning for time series?

Yes, Aeon includes several deep learning architectures specifically for time series, such as InceptionTime, ResNet, and FCN, available for classification and regression.

Aeon Time Series Machine Learning

Name: Aeon Time Series Machine Learning
Author: Yezez9

byYezez9

0•

Data Science & ML

Performs advanced time series machine learning tasks including classification, forecasting, and anomaly detection using a scikit-learn compatible interface.

Aeon is a specialized toolkit designed for temporal data, extending standard machine learning approaches to handle the complexities of sequential patterns. It provides a comprehensive suite of state-of-the-art algorithms for univariate and multivariate time series analysis, covering everything from predicting future values and clustering patterns to identifying unusual subsequences and segmenting change points. This skill is essential for data scientists and developers working with time-indexed observations who require specialized distance metrics, such as Dynamic Time Warping (DTW), and specialized neural architectures beyond traditional ML frameworks.

Key Features

01Extensive library of temporal distance metrics including DTW, ERP, and MSM.

02Pattern discovery tools for motif detection and similarity search.

03Full scikit-learn compatibility for seamless integration into existing ML pipelines.

04Comprehensive time series classification and regression using ROCKET and deep learning.

05Specialized forecasting, anomaly detection, and segmentation for temporal sequences.

060 GitHub stars

Use Cases

01Detecting industrial machinery failures by identifying anomalies in signal patterns.

02Classifying medical waveforms like ECG or EEG data for diagnostic assistance.

03Predicting future trends and values in financial markets or sensor telemetry data.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yezez9/research-agent aeon

For use in Claude.ai and ChatGPT