End-to-End Learning for Partially-Observed Time Series with PyPOTS

2026-04-27

AI Summary

This work addresses the prevalence of partially observed time series (POTS) in real-world scenarios, where existing approaches often decouple missing data handling from downstream tasks, compromising reproducibility and performance. To bridge this gap, we introduce PyPOTS—the first open-source Python framework enabling end-to-end POTS analysis. It unifies five core tasks: missingness simulation, preprocessing, imputation, forecasting, classification, clustering, and anomaly detection, within a modular architecture that offers standardized APIs for both usability and extensibility. The framework supports custom model integration and embedding of domain-specific constraints. Comprehensive benchmarking demonstrates the efficacy of its end-to-end pipeline, and the codebase—accompanied by engineering best-practice guidelines—is publicly released to foster transparent, reproducible time series research and applications.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Partially-observed time series (POTS) is ubiquitous in real-world applications, yet most existing toolchains separate missing-value handling from downstream learning, which limits reproducibility and overall performance. This tutorial introduces PyPOTS, an open-source Python ecosystem for end-to-end data mining and machine learning on POTS. We present practical workflows spanning missingness simulation, data preprocessing, model training, and evaluation across core tasks, including imputation, forecasting, classification, clustering, and anomaly detection. The tutorial consists of two parts: Part I emphasizes hands-on application for practitioners through unified APIs and benchmark-oriented experiments. Part II targets developers and researchers, focusing on extending PyPOTS with custom models, domain-specific constraints, and contribution-ready engineering practices. Participants will gain both conceptual understanding and implementation experience for building robust, transparent, and reusable POTS pipelines in research and production settings. PyPOTS is publicly available at https://github.com/WenjieDu/PyPOTS

Problem

• Partially-Observed Time Series
• missing-value handling
• end-to-end learning
• time series analysis
• reproducibility

Innovation

• end-to-end learning
• partially-observed time series
• missing data imputation
• unified machine learning pipeline
• open-source framework