aeon

Name: aeon
Author: K-Dense-AI

877installs3.1kforks

How to install aeon?

k-dense-ai/scientific-agent-skills/aeon

$npx -y skills add K-Dense-AI/scientific-agent-skills --skill aeon

Installs into the current project.

Summary

This skill should be used for time series machine learning tasks including classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search. Use when working with temporal data, sequential patterns, or time-indexed observations requiring

Files · 1

View on GitHub

SKILL.md

1	# Aeon Time Series Machine Learning
2
3	## Overview
4
5	Aeon is a scikit-learn compatible Python toolkit for time series machine learning ([aeon-toolkit.org](https://www.aeon-toolkit.org/)). It provides algorithms across classification, regression, clustering, forecasting, anomaly detection, segmentation, similarity search, distances, transformations, benchmarking, and visualization — with a consistent estimator API.
6
7	Version note: Examples target aeon 1.x (stable docs: v1.4.0, March 2026). The v1.0 release reworked forecasting and transformations; import paths differ from aeon 0.x/sktime-era code.
8
9	## When to Use This Skill
10
11	Apply this skill when:
12	- Classifying or predicting from time series data
13	- Detecting anomalies or change points in temporal sequences
14	- Clustering similar time series patterns
15	- Forecasting future values
16	- Finding repeated patterns (motifs) or unusual subsequences (discords)
17	- Comparing time series with specialized distance metrics
18	- Extracting features from temporal data
19
20	## Installation
21
22	Requires Python 3.10+ (3.11+ recommended). Pin a 1.x release for reproducibility:
23
24	```bash
25	uv pip install "aeon>=1.4,<2"
26	```
27
28	For deep learning forecasters/classifiers and other optional estimators:
29
30	```bash
31	uv pip install "aeon[all_extras]>=1.4,<2"
32	```
33
34	On zsh, quote the extras: `uv pip install "aeon[all_extras]>=1.4,<2"`.
35
36	### Experimental modules
37
38	Upstream treats forecasting, anomaly_detection, segmentation, similarity_search, and visualisation as experimental — interfaces may change between minor releases. Prefer stable modules (classification, regression, clustering, distances, transformations) for production pipelines unless you need these tasks.
39
40	## Core Capabilities
41
42	### 1. Time Series Classification
43
44	Categorize time series into predefined classes. See `references/classification.md` for complete algorithm catalog.
45
46	Quick Start:
47	```python
48	from aeon.classification.convolution_based import RocketClassifier
49	from aeon.datasets import load_classification
50
51	# Load data
52	X_train, y_train = load_classification("GunPoint", split="train")
53	X_test, y_test = load_classification("GunPoint", split="test")
54
55	# Train classifier
56	clf = RocketClassifier(n_kernels=10000)
57	clf.fit(X_train, y_train)
58	accuracy = clf.score(X_test, y_test)
59	```
60
61	Algorithm Selection:
62	- Speed + Performance: `MiniRocketClassifier`, `Arsenal`
63	- Maximum Accuracy: `HIVECOTEV2`, `InceptionTimeClassifier`
64	- Interpretability: `ShapeletTransformClassifier`, `Catch22Classifier`
65	- Small Datasets: `KNeighborsTimeSeriesClassifier` with DTW distance
66
67	### 2. Time Series Regression
68
69	Predict continuous values from time series. See `references/regression.md` for algorithms.
70
71	Quick Start:
72	```python
73	from aeon.regression.convolution_based import RocketRegressor
74	from aeon.datasets import load_regression
75
76	X_train, y_train = load_regression("Covid3Month", split="train")
77	X_test, y_test = load_regression("Covid3Month", split="test")
78
79	reg = RocketRegressor()
80	reg.fit(X_train, y_train)
81	predictions = reg.predict(X_test)
82	```
83
84	### 3. Time Series Clustering
85
86	Group similar time series without labels. See `references/clustering.md` for methods.
87
88	Quick Start:
89	```python
90	from aeon.clustering import TimeSeriesKMeans
91
92	clusterer = TimeSeriesKMeans(
93	n_clusters=3,
94	distance="dtw",
95	averaging_method="ba"
96	)
97	labels = clusterer.fit_predict(X_train)
98	centers = clusterer.cluster_centers_
99	```
100
101	### 4. Forecasting
102
103	Predict future time series values (experimental module in aeon 1.x). See `references/forecasting.md` for forecasters.
104
105	Quick Start:
106	```python
107	import numpy as np
108	from aeon.forecasting import NaiveForecaster
109	from aeon.forecasting.stats import ARIMA
110
111	y_train = np.array([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0])
112
113	# Set horizon in the constructor; predict passes the series to forecast from
114	naive = NaiveForecaster(strategy="last", horizon=5)
115	naive.fit(y_train)
116	y_pred = naive.predict(y_train)
117
118	# ARIMA uses p/d/q (not order=); multi-step via iterative_forecast
119	arima = ARIMA(p=1, d=1, q=1)
120	arima.fit(y_train)
121	y_pred = arima.iterative_forecast(y_train, prediction_horizon=5)
122	```
123
124	### 5. Anomaly Detection
125
126	Identify unusual patterns or outliers. See `references/anomaly_detection.md

People who liked this also viewed

6 picks

Type

aeon

877installs3.1kforks

View on GitHub

How to install aeon?

k-dense-ai/scientific-agent-skills/aeon

$npx -y skills add K-Dense-AI/scientific-agent-skills --skill aeon

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Aeon Time Series Machine Learning
2
3	## Overview
4
5	Aeon is a scikit-learn compatible Python toolkit for time series machine learning ([aeon-toolkit.org](https://www.aeon-toolkit.org/)). It provides algorithms across classification, regression, clustering, forecasting, anomaly detection, segmentation, similarity search, distances, transformations, benchmarking, and visualization — with a consistent estimator API.
6
7	Version note: Examples target aeon 1.x (stable docs: v1.4.0, March 2026). The v1.0 release reworked forecasting and transformations; import paths differ from aeon 0.x/sktime-era code.
8
9	## When to Use This Skill
10
11	Apply this skill when:
12	- Classifying or predicting from time series data
13	- Detecting anomalies or change points in temporal sequences
14	- Clustering similar time series patterns
15	- Forecasting future values
16	- Finding repeated patterns (motifs) or unusual subsequences (discords)
17	- Comparing time series with specialized distance metrics
18	- Extracting features from temporal data
19
20	## Installation
21
22	Requires Python 3.10+ (3.11+ recommended). Pin a 1.x release for reproducibility:
23
24	```bash
25	uv pip install "aeon>=1.4,<2"
26	```
27
28	For deep learning forecasters/classifiers and other optional estimators:
29
30	```bash
31	uv pip install "aeon[all_extras]>=1.4,<2"
32	```
33
34	On zsh, quote the extras: `uv pip install "aeon[all_extras]>=1.4,<2"`.
35
36	### Experimental modules
37
38	Upstream treats forecasting, anomaly_detection, segmentation, similarity_search, and visualisation as experimental — interfaces may change between minor releases. Prefer stable modules (classification, regression, clustering, distances, transformations) for production pipelines unless you need these tasks.
39
40	## Core Capabilities
41
42	### 1. Time Series Classification
43
44	Categorize time series into predefined classes. See `references/classification.md` for complete algorithm catalog.
45
46	Quick Start:
47	```python
48	from aeon.classification.convolution_based import RocketClassifier
49	from aeon.datasets import load_classification
50
51	# Load data
52	X_train, y_train = load_classification("GunPoint", split="train")
53	X_test, y_test = load_classification("GunPoint", split="test")
54
55	# Train classifier
56	clf = RocketClassifier(n_kernels=10000)
57	clf.fit(X_train, y_train)
58	accuracy = clf.score(X_test, y_test)
59	```
60
61	Algorithm Selection:
62	- Speed + Performance: `MiniRocketClassifier`, `Arsenal`
63	- Maximum Accuracy: `HIVECOTEV2`, `InceptionTimeClassifier`
64	- Interpretability: `ShapeletTransformClassifier`, `Catch22Classifier`
65	- Small Datasets: `KNeighborsTimeSeriesClassifier` with DTW distance
66
67	### 2. Time Series Regression
68
69	Predict continuous values from time series. See `references/regression.md` for algorithms.
70
71	Quick Start:
72	```python
73	from aeon.regression.convolution_based import RocketRegressor
74	from aeon.datasets import load_regression
75
76	X_train, y_train = load_regression("Covid3Month", split="train")
77	X_test, y_test = load_regression("Covid3Month", split="test")
78
79	reg = RocketRegressor()
80	reg.fit(X_train, y_train)
81	predictions = reg.predict(X_test)
82	```
83
84	### 3. Time Series Clustering
85
86	Group similar time series without labels. See `references/clustering.md` for methods.
87
88	Quick Start:
89	```python
90	from aeon.clustering import TimeSeriesKMeans
91
92	clusterer = TimeSeriesKMeans(
93	n_clusters=3,
94	distance="dtw",
95	averaging_method="ba"
96	)
97	labels = clusterer.fit_predict(X_train)
98	centers = clusterer.cluster_centers_
99	```
100
101	### 4. Forecasting
102
103	Predict future time series values (experimental module in aeon 1.x). See `references/forecasting.md` for forecasters.
104
105	Quick Start:
106	```python
107	import numpy as np
108	from aeon.forecasting import NaiveForecaster
109	from aeon.forecasting.stats import ARIMA
110
111	y_train = np.array([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0])
112
113	# Set horizon in the constructor; predict passes the series to forecast from
114	naive = NaiveForecaster(strategy="last", horizon=5)
115	naive.fit(y_train)
116	y_pred = naive.predict(y_train)
117
118	# ARIMA uses p/d/q (not order=); multi-step via iterative_forecast
119	arima = ARIMA(p=1, d=1, q=1)
120	arima.fit(y_train)
121	y_pred = arima.iterative_forecast(y_train, prediction_horizon=5)
122	```
123
124	### 5. Anomaly Detection
125
126	Identify unusual patterns or outliers. See `references/anomaly_detection.md

People who liked this also viewed

6 picks

Type