ermshaua/window-size-selection
This is the supporting website for the paper "Window Size Selection In Unsupervised Time Series Analytics: A Review and Benchmark".
When analyzing time series data for anomalies, patterns, or segments, the choice of 'window size' (how much data you look at at once) significantly impacts the results. This resource provides a systematic comparison of techniques to automatically select the best window size. It takes various time series datasets as input and outputs optimized window sizes, helping researchers and data scientists improve their analysis of time-series data.
Use this if you are a researcher or data scientist needing to understand or reproduce how different window size selection algorithms perform on diverse time series analysis tasks like anomaly detection or segmentation.
Not ideal if you need a plug-and-play Python library for integrating window size selection directly into an application; for that, consider the 'claspy' package instead.
Stars
16
Forks
—
Language
Jupyter Notebook
License
BSD-3-Clause
Category
Last pushed
Feb 27, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ermshaua/window-size-selection"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
aeon-toolkit/aeon
A toolkit for time series machine learning and deep learning
sktime/sktime
A unified framework for machine learning with time series
Nixtla/neuralforecast
Scalable and user friendly neural :brain: forecasting algorithms.
tslearn-team/tslearn
The machine learning toolkit for time series analysis in Python
Nixtla/statsforecast
Lightning ⚡️ fast forecasting with statistical and econometric models.