Update run.py

[v-parmar]

Reference Issue

Documentation Update #1069

What does this PR implement/fix? Explain your changes.

Here, I added a description for each attribute of what it represents.

[mfeurer]

Thank you very much for contribution. This looks very helpful, unfortunately, it contains duplicate information. The type of each variable is now given twice: once in the type hint, and once in the documentation. Could you please update your PR to not contain such duplicate information?

[LennartPurucker]

Heyho @v-parmar,

here is a potential version of what @mfeurer requested.
I also added the type hint for the default parameters.

You could update your PR/Commit with it if everyone thinks this is fine.

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

    Parameters
    ----------
    task_id: int
        The ID of the OpenML task associated with the run.
    flow_id: int
        The ID of the OpenML flow associated with the run.
    dataset_id: int
        The ID of the OpenML dataset used for the run.
    setup_string: str, optional (default=None)
        The setup string of the run.
    output_files: Dict[str, str], optional (default=None)
        The map of file names to file locations. The files are related to the run.
    setup_id: int, optional (default=None)
        The ID of the setup used for the run.
    tags: List[str], optional (default=None)
        The tags associated with the run.
    uploader: int, optional (default=None)
        The user ID of the person who uploaded the run.
    uploader_name: str, optional (default=None)
       The name of the person who uploaded the run.
    evaluations: Dict, optional (default=None)
        The evaluations of the run.
    fold_evaluations: Dict, optional (default=None)
        The evaluations of the run for each fold.
    sample_evaluations: Dict, optional (default=None)
        The run for each sample.
    data_content: List[List], optional (default=None)
        The predictions generated from executing the run.
    trace: OpenMLRunTrace, optional (default=None)
        The trace of the run.
    model: object, optional (default=None)
        The model used to compute the run.
    task_type: str, optional (default=None)
        The type of the OpenML task associated with the run.
    task_evaluation_measure: str, optional (default=None)
        The evaluation measure used for the task.
    flow_name: str, optional (default=None)
        The name of the OpenML flow associated with the run.
    parameter_settings: List[OrderedDict], optional (default=None)
        The parameter settings of the model used for the run.
    predictions_url: str, optional (default=None)
        The URL of the predictions file on the server.
    task: OpenMLTask, optional (default=None)
        The OpenML task object associated with the run.
    flow: OpenMLFlow, optional (default=None)
        The OpenMLFlow object associated with the run.
    run_id: int, optional (default=None)
        The ID of the run.
    description_text: str, optional (default=None)
        The description to be added to the predictions file. If not provided, it is set to the time the arff file was generated.
    run_details: str, optional (default=None)
        Description of the run stored in the run meta-data.
    """

[LennartPurucker]

Hey @v-parmar,

Thank you very much for the update. But I think you only updated some of it. @mfeurer specifically wanted to change the duplicate information. These duplicates still exist in the current version.

If you compare my version from above and the current version, it is mainly about the start of the parameters' descriptions, which usually start with explaining the type. As far as I can tell, this was the problem.

[v-parmar]

Hey @mfeurer, as you said I have updated the description and removed duplicate information. @LennartPurucker Thanks for the hint.