openml.OpenMLRun

class openml.OpenMLRun(task_id, flow_id, dataset_id, setup_string=None, output_files=None, setup_id=None, tags=None, uploader=None, uploader_name=None, evaluations=None, fold_evaluations=None, sample_evaluations=None, data_content=None, trace=None, model=None, task_type=None, task_evaluation_measure=None, flow_name=None, parameter_settings=None, predictions_url=None, task=None, flow=None, run_id=None)

OpenML Run: result of running a model on an openml dataset.

Parameters
task_idint

Refers to the task.

flow_idint

Refers to the flow.

dataset_id: int

Refers to the data.

classmethod from_filesystem(directory: str, expect_model: bool = True) → 'OpenMLRun'

The inverse of the to_filesystem method. Instantiates an OpenMLRun object based on files stored on the file system.

Parameters
directorystr

a path leading to the folder where the results are stored

expect_modelbool

if True, it requires the model pickle to be present, and an error will be thrown if not. Otherwise, the model might or might not be present.

Returns
runOpenMLRun

the re-instantiated run object

get_metric_fn(self, sklearn_fn, kwargs=None)

Calculates metric scores based on predicted values. Assumes the run has been executed locally (and contains run_data). Furthermore, it assumes that the ‘correct’ or ‘truth’ attribute is specified in the arff (which is an optional field, but always the case for openml-python runs)

Parameters
sklearn_fnfunction

a function pointer to a sklearn function that accepts y_true, y_pred and **kwargs

Returns
scoreslist

a list of floats, of length num_folds * num_repeats

publish(self) → 'OpenMLRun'

Publish a run (and if necessary, its flow) to the OpenML server.

Uploads the results of a run to OpenML. If the run is of an unpublished OpenMLFlow, the flow will be uploaded too. Sets the run_id on self.

Returns
selfOpenMLRun
push_tag(self, tag: str) → None

Annotates this run with a tag on the server.

Parameters
tagstr

Tag to attach to the run.

remove_tag(self, tag: str) → None

Removes a tag from this run on the server.

Parameters
tagstr

Tag to attach to the run.

to_filesystem(self, directory: str, store_model: bool = True) → None

The inverse of the from_filesystem method. Serializes a run on the filesystem, to be uploaded later.

Parameters
directorystr

a path leading to the folder where the results will be stored. Should be empty

store_modelbool, optional (default=True)

if True, a model will be pickled as well. As this is the most storage expensive part, it is often desirable to not store the model.