EasyTime: Time Series Forecasting Made Easy

TL;DR

EasyTime tackles the challenges of time series forecasting by integrating a large, diverse benchmark (TFB) with four modular capabilities: one-click evaluation, automated ensemble construction, and natural language Q&A over benchmarking results. It leverages offline representation learning (TS2Vec) and a soft-label loss to rank candidate methods, then builds dataset-specific ensembles online, while NL2SQL-based Q&A enables natural-language queries about method performance. The platform provides a unified, scalable workflow across 8,000+ time series and 30+ methods, with compatibility to existing TSF libraries and diverse evaluation strategies. This design reduces manual effort, accelerates method development, and provides practitioners with intuitive, data-driven tools to select methods and obtain explanations through charts and natural language outputs.

Abstract

Time series forecasting has important applications across diverse domains. EasyTime, the system we demonstrate, facilitates easy use of time-series forecasting methods by researchers and practitioners alike. First, EasyTime enables one-click evaluation, enabling researchers to evaluate new forecasting methods using the suite of diverse time series datasets collected in the preexisting time series forecasting benchmark (TFB). This is achieved by leveraging TFB's flexible and consistent evaluation pipeline. Second, when practitioners must perform forecasting on a new dataset, a nontrivial first step is often to find an appropriate forecasting method. EasyTime provides an Automated Ensemble module that combines the promising forecasting methods to yield superior forecasting accuracy compared to individual methods. Third, EasyTime offers a natural language Q&A module leveraging large language models. Given a question like "Which method is best for long term forecasting on time series with strong seasonality?", EasyTime converts the question into SQL queries on the database of results obtained by TFB and then returns an answer in natural language and charts. By demonstrating EasyTime, we intend to show how it is possible to simplify the use of time series forecasting and to offer better support for the development of new generations of time series forecasting methods.

Figures (5)

• Figure 1: EasyTime Overview.
• Figure 2: Automated Ensemble Overview.
• Figure 3: Workflow of Natural Language Q&A.
• Figure 4: An Example of Method Recommendation, Automated Ensembles, and Forecasts Visualizations.
• Figure 5: An Example of EasyTime Q&A.