⬢github Python · 87 ★ +40 since we first saw it · pushed 21 h ago · MIT
OpenTSLM/TimeNet
Download and explore time-series datasets through one standardized format.
TimeNet is an open-source Python data layer for time-series ML. Datasets ship in one standardized format (TimeF: manifest, DuckDB control store, Parquet/Zarr values), stored as immutable versions in a registry and loaded through a single Python API or CLI. Connectors convert raw sources into TimeF, while training and inference remain in your own framework.
Why now: It was featured on Hacker News as open data infrastructure for temporal foundation models, and an online hackathon around it is currently live.
Who it is for: ML researchers and engineers building time-series forecasting, classification, or foundation models who need consistent datasets.
Stars over our 269 snapshots: 47 to 87, since 2 d ago.
Where people talked about it
API: https://socialmediatrends-api.osmike.com/v1/repos/OpenTSLM/TimeNet