How to use the mlxtend._base._BaseModel function in mlxtend

# Implementation of the mulitnomial logistic regression algorithm for
# classification.

# Author: Sebastian Raschka 
#
# License: BSD 3 clause

import numpy as np
from time import time
from .._base import _BaseModel
from .._base import _IterativeModel
from .._base import _MultiClass
from .._base import _Classifier


class SoftmaxRegression(_BaseModel, _IterativeModel,
                        _Classifier,  _MultiClass):

    """Softmax regression classifier.

    Parameters
    ------------
    eta : float (default: 0.01)
        Learning rate (between 0.0 and 1.0)
    epochs : int (default: 50)
        Passes over the training dataset.
        Prior to each epoch, the dataset is shuffled
        if `minibatches > 1` to prevent cycles in stochastic gradient descent.
    l2 : float
        Regularization parameter for L2 regularization.
        No regularization if l2=0.0.
    minibatches : int (default: 1)

# Sebastian Raschka 2014-2019
# mlxtend Machine Learning Library Extensions
#
# Base Regressor (Regressor Parent Class)
# Author: Sebastian Raschka 
#
# License: BSD 3 clause

import numpy as np
from time import time
from .._base import _BaseModel
from .._base import _IterativeModel
from .._base import _Regressor


class LinearRegression(_BaseModel, _IterativeModel, _Regressor):

    """ Ordinary least squares linear regression.

    Parameters
    ------------
    method : string (default: 'direct')
        For gradient descent-based optimization, use `sgd` (see `minibatch`
        parameter for further options). Otherwise, if `direct` (default),
        the analytical method is used. For alternative, numerically more
        stable solutions, use either `qr` (QR decomopisition) or `svd`
        (Singular Value Decomposition).
    eta : float (default: 0.01)
        solver learning rate (between 0.0 and 1.0). Used with `method =`
        `'sgd'`. (See `methods` parameter for details)
    epochs : int (default: 50)
        Passes over the training dataset.

# Sebastian Raschka 2014-2019
# mlxtend Machine Learning Library Extensions
#
# Principal Component Analysis for dimensionality reduction.
# Author: Sebastian Raschka 
#
# License: BSD 3 clause

import numpy as np
from scipy.spatial import distance
from .._base import _BaseModel


class RBFKernelPCA(_BaseModel):
    """
    RBF Kernel Principal Component Analysis for dimensionality reduction.

    Parameters
    ----------
    gamma : float (default: 15.0)
        Free parameter (coefficient) of the RBF kernel.
    n_components : int (default: None)
        The number of principal components for transformation.
        Keeps the original dimensions of the dataset if `None`.
    copy_X : bool (default: True)
        Copies training data, which is required to compute the projection
        of new data via the transform method. Uses a reference to X if False.

    Attributes
    ----------

# Implementation of a Multi-layer Perceptron in Tensorflow
# Author: Sebastian Raschka 
#
# License: BSD 3 clause

import tensorflow as tf
import numpy as np
from time import time
from .._base import _BaseModel
from .._base import _IterativeModel
from .._base import _MultiClass
from .._base import _MultiLayer
from .._base import _Classifier


class TfMultiLayerPerceptron(_BaseModel, _IterativeModel,
                             _MultiClass, _MultiLayer, _Classifier):
    """Multi-layer perceptron classifier.

    Parameters
    ------------
    eta : float (default: 0.5)
        Learning rate (between 0.0 and 1.0)
    epochs : int (default: 50)
        Passes over the training dataset.
        Prior to each epoch, the dataset is shuffled
        if `minibatches > 1` to prevent cycles in stochastic gradient descent.
    hidden_layers : list (default: [50, 10])
        Number of units per hidden layer. By default 50 units in the
        first hidden layer, and 10 hidden units in the second hidden layer.
    n_classes : int (default: None)
        A positive integer to declare the number of class labels