Skip to content

Client

timenet.client source

The :class:TimeNet SDK: the single entry point for using TimeNet from code.

Wraps a :class:~timenet.registry.BaseRegistry (the catalog) and a local storage path (the download cache), exposing list / get / search / download / load. It never runs connector code; producing datasets is the curation side.

T module-attribute source

T = TypeVar('T')

TimeNet source

Browse a registry and fetch datasets to local storage.

download source

download(
    dataset_id: str,
    version: str | None = None,
    *,
    force: bool = False,
) -> Path

Fetch a dataset version's files into local storage and return its directory.

Parameters:

Name Type Description Default
dataset_id str

The dataset id.

 required
version str | None

The version string, or None for the latest.

 None
force bool

Re-download even if an up-to-date copy already exists.

 False

Returns:

Type Description
Path

The local <storage>/<dataset_id>/<version>/ directory.

get source

get(
    dataset_id: str, version: str | None = None
) -> Manifest

Return a dataset's manifest.

Parameters:

Name Type Description Default
dataset_id str

The dataset id.

 required
version str | None

The version string, or None for the latest.

 None

Returns:

Type Description
Manifest

The dataset's manifest.

list source

list() -> _Metadatas

Return the metadata of every dataset in the registry.

Returns:

Name Type Description
One _Metadatas

class:~timenet.types.DatasetMetadata per dataset.

load source

load(
    dataset_id: str, version: str | None = None
) -> TimeFDataset

Download if needed, then read the dataset into memory.

Parameters:

Name Type Description Default
dataset_id str

The dataset id.

 required
version str | None

The version string, or None for the latest.

 None

Returns:

Type Description
TimeFDataset

The dataset with lazy per-series loaders backed by the local copy.

load_torch source

load_torch(
    dataset_id: str, version: str | None = None
) -> TimeFTorchDataset

Download if needed and return the dataset as a read-only PyTorch Dataset.

Requires the torch extra (install coming soon); the torch view is imported lazily so base users don't need torch.

Parameters:

Name Type Description Default
dataset_id str

The dataset id.

 required
version str | None

The version string, or None for the latest.

 None

Returns:

Name Type Description
A TimeFTorchDataset

class:~timenet.torch.TimeFTorchDataset over the loaded dataset.

search source

search(
    *,
    query: _OrList[str] = None,
    domain: _OrList[Domain] = None,
    task: _OrList[type[Task]] = None,
    license: _OrList[License] = None,
    time_series_spec: _OrList[str] = None,
    dataset_id: _OrList[str] = None,
    tag: _OrList[str] = None,
    limit: int = 100,
) -> _Metadatas

Search the registry. Mirrors :meth:~timenet.registry.BaseRegistry.search.

Parameters:

Name Type Description Default
query _OrList[str]

Free-text terms over name/description/tags.

 None
domain _OrList[Domain]

Keep datasets sharing any of these domains.

 None
task _OrList[type[Task]]

Keep datasets whose schema includes any of these task classes.

 None
license _OrList[License]

Keep datasets with any of these licenses.

 None
time_series_spec _OrList[str]

Keep datasets declaring all of these spec_type values.

 None
dataset_id _OrList[str]

Keep only these ids.

 None
tag _OrList[str]

Keep datasets declaring all of these tags.

 None
limit int

Maximum number of results.

 100

Returns:

Type Description
_Metadatas

The matching dataset metadata.