How to use the lime.lime_base function in lime

if(self.training_data_stats is None):
                discretized_training_data = self.discretizer.discretize(
                    training_data)

        if kernel_width is None:
            kernel_width = np.sqrt(training_data.shape[1]) * .75
        kernel_width = float(kernel_width)

        if kernel is None:
            def kernel(d, kernel_width):
                return np.sqrt(np.exp(-(d ** 2) / kernel_width ** 2))

        kernel_fn = partial(kernel, kernel_width=kernel_width)

        self.feature_selection = feature_selection
        self.base = lime_base.LimeBase(kernel_fn, verbose, random_state=self.random_state)
        self.class_names = class_names

        # Though set has no role to play if training data stats are provided
        self.scaler = sklearn.preprocessing.StandardScaler(with_mean=False)
        self.scaler.fit(training_data)
        self.feature_values = {}
        self.feature_frequencies = {}

        for feature in self.categorical_features:
            if training_data_stats is None:
                if self.discretizer is not None:
                    column = discretized_training_data[:, feature]
                else:
                    column = training_data[:, feature]

                feature_count = collections.Counter(column)

otherwise.
            random_state: an integer or numpy.RandomState that will be used to
                generate random numbers. If None, the random state will be
                initialized using the internal numpy seed.
            char_level: an boolean identifying that we treat each character
                as an independent occurence in the string
        """

        if kernel is None:
            def kernel(d, kernel_width):
                return np.sqrt(np.exp(-(d ** 2) / kernel_width ** 2))

        kernel_fn = partial(kernel, kernel_width=kernel_width)

        self.random_state = check_random_state(random_state)
        self.base = lime_base.LimeBase(kernel_fn, verbose,
                                       random_state=self.random_state)
        self.class_names = class_names
        self.vocabulary = None
        self.feature_selection = feature_selection
        self.bow = bow
        self.mask_string = mask_string
        self.split_expression = split_expression
        self.char_level = char_level

details on what each of the options does.
            random_state: an integer or numpy.RandomState that will be used to
                generate random numbers. If None, the random state will be
                initialized using the internal numpy seed.
        """
        kernel_width = float(kernel_width)

        if kernel is None:
            def kernel(d, kernel_width):
                return np.sqrt(np.exp(-(d ** 2) / kernel_width ** 2))

        kernel_fn = partial(kernel, kernel_width=kernel_width)

        self.random_state = check_random_state(random_state)
        self.feature_selection = feature_selection
        self.base = lime_base.LimeBase(kernel_fn, verbose, random_state=self.random_state)

from __future__ import unicode_literals

import itertools,re,copy,types,os
import numpy as np
import scipy as sp
import sklearn
import skimage as si
from skimage import io,segmentation,util

import streakFunctions as streak
from time import time
from lime import explanation
from lime import lime_base
from lime import lime_image

class MyLimeBase(lime_base.LimeBase):
    '''
    Extend LimeBase class with a new feature selection method that supports streaming_greedy feature selection
    This is described in Section A.8 of the paper
    '''
    def feature_selection(self, data, labels, weights, num_features, method):
        """Selects features for the model. see explain_instance_with_data to
           understand the parameters."""
        tstart = time()
        # print "method is %s" % method
        if method in ['thresholded_greedy','streaming_greedy','streaming_regression','streaming','STREAK','streak','streak_lime']:
            out = self.thresholded_greedy(data,labels,weights,num_features)
            tend = time()
            self.selection_time = tend - tstart
            return out
        else:
            # print "Using LimeBase.feature_selection"