import argparse
import sklearn
import sklearn.datasets
import sklearn.pipeline
import sklearn.model_selection
import sklearn.linear_model

parser = argparse.ArgumentParser(description='Playground for sklearn models')
parser.add_argument('--seed', type=int, default=42, help='random seed')
parser.add_argument('--test_size', type=float, default=0.2, help='float indicating the ratio of the test examples to the whole dataset')

def fetch_dataset(args):
    return sklearn.datasets.fetch_california_housing()

def create_model(args):
    model = sklearn.pipeline.Pipeline([
        ('linear_regression', sklearn.linear_model.SGDClassifier(verbose=1, random_state=args.seed))
    ])

    return model

if __name__ == "__main__":
    args = parser.parse_args()
    dataset = sklearn.datasets.load_breast_cancer()
    train_data, test_data, train_target, test_target = sklearn.model_selection.train_test_split(dataset.data, dataset.target, test_size=args.test_size, random_state=args.seed)

    # create the model
    # this is your main playground
    model = create_model(args)

    # fit (train) the model on the training data
    model.fit(train_data, train_target)

    # predict on the test set
    prediction = model.predict(test_data)
    accuracy = sklearn.metrics.accuracy_score(test_target, prediction)
    print(accuracy)