How to use the scikit-learn.sklearn.base.BaseEstimator function in scikit-learn

from .sgd_fast import ModifiedHuber
from .sgd_fast import SquaredLoss
from .sgd_fast import Huber
from .sgd_fast import EpsilonInsensitive
from .sgd_fast import SquaredEpsilonInsensitive

LEARNING_RATE_TYPES = {"constant": 1, "optimal": 2, "invscaling": 3,
                       "pa1": 4, "pa2": 5}

PENALTY_TYPES = {"none": 0, "l2": 2, "l1": 1, "elasticnet": 3}

DEFAULT_EPSILON = 0.1
# Default value of ``epsilon`` parameter.


class BaseSGD(six.with_metaclass(ABCMeta, BaseEstimator, SparseCoefMixin)):
    """Base class for SGD classification and regression."""

    def __init__(self, loss, penalty='l2', alpha=0.0001, C=1.0,
                 l1_ratio=0.15, fit_intercept=True, n_iter=5, shuffle=True,
                 verbose=0, epsilon=0.1, random_state=None,
                 learning_rate="optimal", eta0=0.0, power_t=0.5,
                 warm_start=False, average=False):
        self.loss = loss
        self.penalty = penalty
        self.learning_rate = learning_rate
        self.epsilon = epsilon
        self.alpha = alpha
        self.C = C
        self.l1_ratio = l1_ratio
        self.fit_intercept = fit_intercept
        self.n_iter = n_iter