High Level API doctestable

from dffml import Features, DefFeature
from dffml.noasync import train, accuracy, predict
from dffml_model_scikit import LinearRegressionModel

def highLevelApi_example(feature_def, pred_feature_def, train_data, accuracy_data, prediction_data):


    """
    .. doctest::

    >>> highLevelApi_example(Features(
    ...       DefFeature("Years", int, 1),
    ...       DefFeature("Expertise", int, 1),
    ...       DefFeature("Trust", float, 1),
    ...       ), DefFeature("Salary", int, 1), 
    ...       [
    ...       {"Years": 0, "Expertise": 1, "Trust": 0.2, "Salary": 10},
    ...       {"Years": 1, "Expertise": 3, "Trust": 0.4, "Salary": 20},
    ...       {"Years": 2, "Expertise": 5, "Trust": 0.6, "Salary": 30},
    ...       {"Years": 3, "Expertise": 7, "Trust": 0.8, "Salary": 40},
    ...       ],
    ...       [
    ...       {"Years": 4, "Expertise": 9, "Trust": 1.0, "Salary": 50},
    ...       {"Years": 5, "Expertise": 11, "Trust": 1.2, "Salary": 60},
    ...       ],
    ...       [
    ...       {"Years": 6, "Expertise": 13, "Trust": 1.4},
    ...       {"Years": 7, "Expertise": 15, "Trust": 1.6},
    ...       ]
    ...       )
    Accuracy: 1.0
    {'Salary': {'confidence': 1.0, 'value': 70.0}}
    {'Salary': {'confidence': 1.0, 'value': 80.0}}
    """

    ## Creating a model

    model = LinearRegressionModel(
    features= feature_def
    , predict=pred_feature_def
    )

    ## train_data, accuracy_data and prediction_data are lists of dictionaries here.

    ## Training our model

    train(model, *train_data)

    ## Checking our model's accuracy

    print("Accuracy:", accuracy(model, *accuracy_data,))

    ## Making predictions on on new data

    for i, features, prediction in predict(
    model,
    *prediction_data
    ):
        print(prediction)