def get_cls_model(input_shape):
    """
    :param input_shape: tuple (n_rows, n_cols, n_channgels)
            input shape of image for classification
    :return: nn model for classification
    """
    return Sequential([
        Conv2D(64, (3, 3), input_shape=input_shape, activation='relu', padding="same"),
        MaxPooling2D(pool_size=(2, 2)),

        Conv2D(128, (3, 3), activation='relu', padding="same"),
        Conv2D(128, (3, 3), activation='relu', padding="same"),
        MaxPooling2D(pool_size=(2, 2)),

        # Conv2D(256, (3, 3), activation='relu', padding="same"),
        # Conv2D(256, (3, 3), activation='relu', padding="same"),
        # Conv2D(256, (3, 3), activation='relu', padding="same"),
        # MaxPooling2D(pool_size=(2, 2)),


        # Conv2D(512, (3, 3), activation='relu'),
        # Conv2D(512, (3, 3), activation='relu'),
        # MaxPooling2D(pool_size=(2, 2)),

        # Conv2D(512, (3, 3), activation='relu'),
        # MaxPooling2D(pool_size=(2, 2)),

        Flatten(),
        Dense(64, activation='relu'),
        Dense(2),
        Activation('softmax')
    ])

def fit_cls_model(X, y, force_save=False):
    """
    :param X: 4-dim ndarray with training images
    :param y: 2-dim ndarray with one-hot labels for training
    :return: trained nn model
    """
    model = get_cls_model((40, 100, 1))
    adam = optimizers.Adam(learning_rate=0.0001)
    model.compile(loss='categorical_crossentropy', optimizer=adam,
                    metrics=[metrics.categorical_accuracy])
    print("model is compiled")
    model.fit(x=X, y=y, epochs=5)
    if (force_save):
        model.save('classifier_model.h5')

    # print("\n\n\n\n\n\n")
    print(X.shape, y.shape)
    return model