How to use the ramp-database.ramp_database.tools._query.select_submissions_by_state function in ramp-database

def delete_event(session, event_name):
    """Delete an event from the database.

    Parameters
    ----------
    session : :class:`sqlalchemy.orm.Session`
        The session to directly perform the operation on the database.
    event_name : str
        The name of the event to delete.
    """
    event = select_event_by_name(session, event_name)
    submissions = select_submissions_by_state(session, event_name, state=None)
    for sub in submissions:
        delete_submission_similarity(session, sub.id)
    session.delete(event)
    session.commit()

List of submissions information. Each item is a tuple containing:

        * an integer containing the id of the submission;
        * a string with the name of the submission in the database;
        * a list of string representing the file associated with the
          submission.

    See also
    --------
    ramp_database.tools.get_submission_by_id : Get a submission using an id.
    ramp_database.tools.get_submission_by_name : Get a submission using names.
    """
    if state is not None and state not in STATES:
        raise UnknownStateError("Unrecognized state : '{}'".format(state))

    submissions = select_submissions_by_state(session, event_name, state)

    if not submissions:
        return []

    submission_id = [sub.id for sub in submissions]
    submission_files = [sub.files for sub in submissions]
    submission_basename = [sub.basename for sub in submissions]
    filenames = [[f.path for f in files] for files in submission_files]
    return list(zip(submission_id, submission_basename, filenames))

def compute_historical_contributivity(session, event_name):
    """Compute historical contributivities of an event using
       contributivities from blending and credits.

    Parameters
    ----------
    session : :class:`sqlalchemy.orm.Session`
        The session to directly perform the operation on the database.
    event_name : str
        The event associated to the submission.
    """
    submissions = select_submissions_by_state(
        session, event_name, state='scored')
    submissions.sort(key=lambda x: x.submission_timestamp, reverse=True)
    for s in submissions:
        s.historical_contributivity = 0.0
    for s in submissions:
        s.historical_contributivity += s.contributivity
        similarities = session.query(SubmissionSimilarity).filter_by(
            type='target_credit', target_submission=s).all()
        if similarities:
            # if a target team enters several credits to a source submission
            # we only take the latest
            similarities.sort(key=lambda x: x.timestamp, reverse=True)
            processed_submissions = []
            historical_contributivity = s.historical_contributivity
            for ss in similarities:
                source_submission = ss.source_submission

The event associated to the submission.
    ramp_kit_dir : str
        The directory of the RAMP kit.
    ramp_data_dir : str
        The directory of the data.
    min_improvement : float, default is 0.0
        The minimum improvement under which greedy blender is stopped.
    """
    logging.basicConfig(level=logging.INFO, format='%(message)s')
    logger.info('Combining models')

    event = select_event_by_name(session, event_name)
    ramp_submission_dir = event.path_ramp_submissions
    score_type = event.get_official_score_type(session)

    submissions = select_submissions_by_state(
        session, event_name, state='scored')
    if len(submissions) == 0:
        logger.info('No submissions to blend.')
        return
    
    blend_submissions(
            submissions=[sub.basename for sub in submissions],
            ramp_kit_dir=ramp_kit_dir,
            ramp_data_dir=ramp_data_dir,
            ramp_submission_dir=ramp_submission_dir,
            save_output=True,
            min_improvement=min_improvement,
    )

    bsc_f_name = 'bagged_scores_combined.csv'
    bsc_df = pd.read_csv(