TSFM-ScalingLaws-Dataset

TSFM-ScalingLaws-Dataset

This is the dataset for the paper Towards Neural Scaling Laws for Time Series Foundation Models.

We selected some high-quality (SNR > 20), domain-balanced, and low-redundancy data from Lotsa and UTS datasets for TSFM pre-training. These datasets contain three sizes: 16B, 1B, 100M, and 10M.

Code: https://github.com/Qingrenn/TSFM-ScalingLaws

Well-trained models: https://huggingface.co/PeacefulData/TSFM-ScalingLaws-Checkpoints