Train scikit-learn compatible time series models for classification, forecasting, and anomaly detection.
npx skills add https://github.com/CRAG666/dotfiles --skill aeon-crag666
Please help me install this Agent Skill. Skill: aeon Source: https://github.com/CRAG666/dotfiles/tree/main/skills/aeon Command: npx skills add https://github.com/CRAG666/dotfiles --skill aeon-crag666
Aeon helps you analyze time-indexed data by providing end-to-end tools to learn from sequences, forecast future values, detect anomalies, and discover similar or repeated patterns—without you having to wire up many separate ML components manually.
Train a time series classifier on a benchmark dataset by running: from aeon.classification.convolution_based import RocketClassifier; from aeon.datasets import load_classification; X_train, y_train = load_classification("GunPoint", split="train"); X_test, y_test = load_classification("GunPoint", split="test"); clf = RocketClassifier(n_kernels=10000); clf.fit(X_train, y_train); print(clf.score(X_test, y_test))
High-intent search queries and answers about installing and using this skill.
To perform time series forecasting and classification, you can use the scikit-learn compatible workflow to train models like RocketClassifier or ARIMA on univariate and multivariate temporal data for predictions.
Anomaly detection and similarity search in temporal data use matrix-profile methods to find motifs, subsequence neighbors, and discord patterns via specialized distance functions without wiring separate components manually.
Yes, the workflow satisfies scikit-learn compatibility, allowing you to apply time series feature extraction like ROCKET variants and normalization directly within standard model training and evaluation pipelines.
Yes, it supports both univariate and multivariate time-indexed sequences for machine learning tasks including classification, regression, clustering, forecasting, segmentation, and similarity search.
Finding motifs and discords in a time series dataset uses matrix-profile discovery algorithms and distance-metric-based methods to identify repeated patterns and anomalies in the temporal sequences.
Available algorithms for time series forecasting and classification include ROCKET variants, DTW-based approaches, and ARIMA/ETS/TANS-based forecasters applied directly to time-series shaped inputs.
LeonChaoXBuild scikit-learn compatible time-series models for classification, regression, clustering, forecasting, and anomaly detection.
ovachieverProvide scikit-learn compatible time series machine learning algorithms for classification, regression, clustering, forecasting, and anomaly detection.
yf8578Perform time series classification, forecasting, and anomaly detection with scikit-learn compatible APIs.
jackspaceSolve time-series machine learning tasks with a scikit-learn-compatible Python API.
