Module `minder_utils.models.feature_selectors.unsupervised.filter`

Expand source code

from sklearn.feature_selection import VarianceThreshold
from minder_utils.models.utils import Feature_selector


class Unsupervised_Filter(Feature_selector):
    '''
    This class provide a set of unsupervised feature selection methods.

    Currently, it contains:
        - VarianceThreshold

    ```Example```
    ```
    from minder_utils.models.feature_selectors.unsupervised.filter import Unsupervised_Filter

    selector = Unsupervised_Filter(model='vt')
    # show the available methods:
    selector.get_info(verbose=True)

    # train the selector. Note the X is the data, y is None and will not be used
    selector.fit(X, y)

    # do the selection
    X = selector.transform(X)
    ```
    '''
    def __init__(self, model_name='vt'):
        super().__init__(model_name)

    @property
    def methods(self):
        return {
            'vt': 'VarianceThreshold',
        }

    @staticmethod
    def vt():
        return VarianceThreshold()

    def __name__(self):
        return 'Unsupervised Filter', self.name

    def fit(self, X, y=None):
        return self.model.fit(X)

    def transform(self, X):
        return self.model.transform(X)

Classes

class Unsupervised_Filter (model_name='vt')

This class provide a set of unsupervised feature selection methods.

Currently, it contains: - VarianceThreshold

Example

from minder_utils.models.feature_selectors.unsupervised.filter import Unsupervised_Filter

selector = Unsupervised_Filter(model='vt')
# show the available methods:
selector.get_info(verbose=True)

# train the selector. Note the X is the data, y is None and will not be used
selector.fit(X, y)

# do the selection
X = selector.transform(X)

Expand source code

class Unsupervised_Filter(Feature_selector):
    '''
    This class provide a set of unsupervised feature selection methods.

    Currently, it contains:
        - VarianceThreshold

    ```Example```
    ```
    from minder_utils.models.feature_selectors.unsupervised.filter import Unsupervised_Filter

    selector = Unsupervised_Filter(model='vt')
    # show the available methods:
    selector.get_info(verbose=True)

    # train the selector. Note the X is the data, y is None and will not be used
    selector.fit(X, y)

    # do the selection
    X = selector.transform(X)
    ```
    '''
    def __init__(self, model_name='vt'):
        super().__init__(model_name)

    @property
    def methods(self):
        return {
            'vt': 'VarianceThreshold',
        }

    @staticmethod
    def vt():
        return VarianceThreshold()

    def __name__(self):
        return 'Unsupervised Filter', self.name

    def fit(self, X, y=None):
        return self.model.fit(X)

    def transform(self, X):
        return self.model.transform(X)

Ancestors

Feature_selector
abc.ABC

Static methods

def vt()

Expand source code

@staticmethod
def vt():
    return VarianceThreshold()

Instance variables

var methods

Expand source code

@property
def methods(self):
    return {
        'vt': 'VarianceThreshold',
    }

Methods

def fit(self, X, y=None)

Expand source code

def fit(self, X, y=None):
    return self.model.fit(X)

def transform(self, X)

Expand source code

def transform(self, X):
    return self.model.transform(X)