data, y = utils.load_data()

    # One hot encoding of target vector
    Y = pd.get_dummies(y).values
    n_classes = Y.shape[1]

    for nodes in [300]:
        for key, X in data.items():
            print('Running : ', key, nodes, X.shape)

            # Split data into training and testing data
            X_train, X_test, Y_train, Y_test = train_test_split(
                X, Y, test_size=0.2, random_state=1, stratify=y)

            # Normalize data with mean 0 and std 1
            X_scaled = normalize(X_train)

            # Train the neural network model
            n_features = X.shape[1]
            model = nn_model(n_features, n_classes, nodes, 0.2)
            history = model.fit(X_scaled, Y_train,
                                epochs=100,
                                batch_size=5,
                                verbose=1,
                                callbacks=[csv_logger])