Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

sktime: A Unified Interface for Machine Learning with Time Series

About

We present sktime -- a new scikit-learn compatible Python library with a unified interface for machine learning with time series. Time series data gives rise to various distinct but closely related learning tasks, such as forecasting and time series classification, many of which can be solved by reducing them to related simpler tasks. We discuss the main rationale for creating a unified interface, including reduction, as well as the design of sktime's core API, supported by a clear overview of common time series tasks and reduction approaches.

Markus L\"oning, Anthony Bagnall, Sajaysurya Ganesh, Viktor Kazakov, Jason Lines, Franz J. Kir\'aly• 2019

Related benchmarks

TaskDatasetResultRank
Time-series classificationCHARACTER TRAJ. (test)
Accuracy0.9
88
Time-series classificationJapanese Vowels (test)
Accuracy96.8
31
Time Series ForecastingAustralian Retail Turnover
MAPE10.734
19
Time Series ForecastingAustralian Electricity Demand
MAPE6.208
19
Time-series classificationAtrialFibrillation (AF) (test)
Accuracy33.3
15
Time-series classificationStandWalkJump (SWJ) (test)
Accuracy46.7
15
Time-series classificationSpokenArabicDigits (SAD) (test)
Accuracy52.4
13
Time-series classificationShakGWZ (test)
Accuracy68
12
Time-series classificationAllGeWX (test)
Accuracy27.3
12
Time-series classificationGPebbleZ1 (test)
Accuracy59.9
12
Showing 10 of 32 rows

Other info

Follow for update