TIME

We present TIME, a task-centric benchmark comprising 50 fresh datasets and 98 forecasting tasks, where configurations are aligned with real-world operational requirements. By archiving window-level prediction results, our benchmark enables pattern-level evaluation and visualization. This repository contains the ready-to-use Hugging Face dataset for running experiments.

🏆Leaderboard

📂Processed csv (Optional: For reference or custom pipelines only)

📊Reuslts & tsfeatures (Optional: Leaderboard data, not required for experiments)

Update Log

• 2026-05-25: Update the dataset license to CC BY-NC 4.0 to ensure compliance with all constituent data providers.
• 2026-05-20: Update Crypto/D. Update data with public market aggregates from BitInfoCharts.
• 2026-05-16: Update Global_Influenza/W. Fix the bug reported in Issue #5.
• 2026-02-10: Initial Release. Introduced a comprehensive benchmark comprising 50 datasets across 98 evaluation tasks.