Original Results Predictor

def predict_results(event):
    """
    Takes an event as a string parameter and uses a decision tree regressor
    to predict the top 3 results given medal and nationality of a competitor
    """

    filtered = df_converted[df_converted['Event'] == event]
    filtered = filtered.loc[:, ['Medal', 'Nationality', 'Result']]
    X = filtered.loc[:, filtered.columns != 'Result']
    y = filtered['Result']
    X = pd.get_dummies(X)

    (X_train, X_test, y_train, y_test) = train_test_split(X, y,
                                                          test_size=0.35)
    model = DecisionTreeRegressor()

    model.fit(X_train, y_train)
    y_test_pred = model.predict(X_test)
    ordered = sorted(y_test_pred.tolist())

    if event in field:

        events.append(event)
        medals.append('Gold')
        mark.append(convert_from_seconds(ordered[2]))

        events.append(event)
        medals.append('Silver')
        mark.append(convert_from_seconds(ordered[1]))

        events.append(event)
        medals.append('Bronze')
        mark.append(convert_from_seconds(ordered[0]))
    else:

        events.append(event)
        medals.append('Gold')
        mark.append(convert_from_seconds(ordered[0]))

        events.append(event)
        medals.append('Silver')
        mark.append(convert_from_seconds(ordered[1]))

        events.append(event)
        medals.append('Bronze')
        mark.append(convert_from_seconds(ordered[2]))


def simulate():
    """
    Predicts the top 3 results for every Olympic event
    Returns a dataframe containing the event name, medal type, and mark of
    each result
    """
    event_array = pd.unique(df_converted['Event'])

    for event in event_array:
        predict_results(event)

    return pd.DataFrame({'Event': events, 'Medal': medals,
                        'Result': mark})