Concepts¶
The vocabulary that shows up across these docs, in one place. Follow a link for the full treatment.
| Term | What it is |
|---|---|
| TimeF | The shared format every dataset is stored and loaded in: one on-disk shape and one API, so an ECG, an accelerometer trace, and a market series all read the same way. |
| Dataset | A named, versioned collection of samples in TimeF, addressed as org/name (for example chengsenwang/tsqa). |
| Sample | One record in a dataset (say, a single patient recording): its time series plus tasks, annotations, and metadata. |
| Time series | One channel's float32 values over time within a sample. Values are read on demand with to_arrow() or to_numpy(). |
| Manifest | The compiled manifest.json for a dataset version: the card's metadata plus the schema derived from the data. The single source of truth the SDK reads. |
| Connector | One BaseConnector per dataset. download() fetches the raw source, convert() builds a TimeFDataset. It knows nothing about the engine or registry. |
| Engine | run_pipeline: drives any connector through the fixed download -> convert -> derive_schema -> store pipeline, and owns caching and idempotency. |
| Curation | Running a connector through the engine to compile a dataset and publish it to a registry, via the timenet-curate CLI. |
| Registry | A served location of compiled manifests and parquet that the SDK reads. It never runs connector code: a local directory, S3, or a remote host. |
| Client (SDK) | TimeNet: the consumer entry point to search, download, and load datasets. |
| Version | An immutable snapshot of a dataset, addressed org/name@version. It is committed atomically once its manifest.json lands. |