kedro.framework.session.session.KedroSession

class kedro.framework.session.session.KedroSession(session_id, package_name=None, project_path=None, save_on_close=False)

KedroSession is the object that is responsible for managing the lifecycle of a Kedro run. - Use KedroSession.create(“<your-kedro-project-package-name>”) as a context manager to construct a new KedroSession with session data provided (see the example below). - Use KedroSession(session_id=<id>) to instantiate an existing session with a given ID.

Example:

from kedro.framework.session import KedroSession

with KedroSession.create("<your-kedro-project-package-name>") as session:
    session.run()

Attributes

store

Return a copy of internal store.

Methods

close()	Close the current session and save its store to disk if save_on_close attribute is True.
create([package_name, project_path, …])	Create a new instance of KedroSession with the session data.
load_context()	An instance of the project context.
run([pipeline_name, tags, runner, …])	Runs the pipeline with a specified runner.

close()

Close the current session and save its store to disk if save_on_close attribute is True.

classmethod create(package_name=None, project_path=None, save_on_close=True, env=None, extra_params=None)

Create a new instance of KedroSession with the session data.

Parameters

• package_name (Optional[str]) – Package name for the Kedro project the session is created for.

• project_path (Union[Path, str, None]) – Path to the project root directory. Default is current working directory Path.cwd().

• save_on_close (bool) – Whether or not to save the session when it’s closed.

• env (Optional[str]) – Environment for the KedroContext.

• extra_params (Optional[Dict[str, Any]]) – Optional dictionary containing extra project parameters for underlying KedroContext. If specified, will update (and therefore take precedence over) the parameters retrieved from the project configuration.

Return type

KedroSession

Returns

A new KedroSession instance.

load_context()

An instance of the project context.

Return type

KedroContext

run(pipeline_name=None, tags=None, runner=None, node_names=None, from_nodes=None, to_nodes=None, from_inputs=None, to_outputs=None, load_versions=None)

Runs the pipeline with a specified runner.

Parameters

• pipeline_name (Optional[str]) – Name of the pipeline that is being run.

• tags (Optional[Iterable[str]]) – An optional list of node tags which should be used to filter the nodes of the Pipeline. If specified, only the nodes containing any of these tags will be run.

• runner (Optional[AbstractRunner]) – An optional parameter specifying the runner that you want to run the pipeline with.

• node_names (Optional[Iterable[str]]) – An optional list of node names which should be used to filter the nodes of the Pipeline. If specified, only the nodes with these names will be run.

• from_nodes (Optional[Iterable[str]]) – An optional list of node names which should be used as a starting point of the new Pipeline.

• to_nodes (Optional[Iterable[str]]) – An optional list of node names which should be used as an end point of the new Pipeline.

• from_inputs (Optional[Iterable[str]]) – An optional list of input datasets which should be used as a starting point of the new Pipeline.

• to_outputs (Optional[Iterable[str]]) – An optional list of output datasets which should be used as an end point of the new Pipeline.

• load_versions (Optional[Dict[str, str]]) – An optional flag to specify a particular dataset version timestamp to load.

Raises

• KedroContextError – If the named or __default__ pipeline is not defined by register_pipelines.

• Exception – Any uncaught exception during the run will be re-raised after being passed to on_pipeline_error hook.

Return type

Dict[str, Any]

Returns

Any node outputs that cannot be processed by the DataCatalog. These are returned in a dictionary, where the keys are defined by the node outputs.

property store

Return a copy of internal store.

Return type

Dict[str, Any]