niaaml.preprocessing

class niaaml.preprocessing.PreprocessingAlgorithm(**kwargs)

Bases: niaaml.pipeline_component.PipelineComponent

Class for implementing preprocessing algorithms.

Date:

2020

Author:

Luka Pečnik

License:

MIT

See Also:

• niaaml.pipeline_component.PipelineComponent

niaaml.preprocessing.feature_selection

class niaaml.preprocessing.feature_selection.BatAlgorithm(**kwargs)

Bases: niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Implementation of feature selection using BA algorithm.

Date:

2020

Author:

Luka Pečnik

Reference:

The implementation is adapted according to the following article: D. Fister, I. Fister, T. Jagrič, I. Fister Jr., J. Brest. A novel self-adaptive differential evolution for feature selection using threshold mechanism . In: Proceedings of the 2018 IEEE Symposium on Computational Intelligence (SSCI 2018), pp. 17-24, 2018.

Reference URL:

http://iztok-jr-fister.eu/static/publications/236.pdf

License:

MIT

See Also:

• niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Name = 'Bat Algorithm'

select_features(x, y, **kwargs)

Perform the feature selection process.

Arguments:

x (pandas.core.frame.DataFrame): Array of original features. y (pandas.core.series.Series) Expected classifier results.

Returns:

pandas.core.frame.DataFrame: Mask of selected features.

set_parameters(**kwargs)

Set the parameters/arguments of the algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

class niaaml.preprocessing.feature_selection.DifferentialEvolution(**kwargs)

Bases: niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Implementation of feature selection using DE algorithm.

Date:

2020

Author:

Luka Pečnik

Reference:

The implementation is adapted according to the following article: D. Fister, I. Fister, T. Jagrič, I. Fister Jr., J. Brest. A novel self-adaptive differential evolution for feature selection using threshold mechanism . In: Proceedings of the 2018 IEEE Symposium on Computational Intelligence (SSCI 2018), pp. 17-24, 2018.

Reference URL:

http://iztok-jr-fister.eu/static/publications/236.pdf

License:

MIT

See Also:

• niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Name = 'Differential Evolution'

select_features(x, y, **kwargs)

Perform the feature selection process.

Arguments:

x (pandas.core.frame.DataFrame): Array of original features. y (pandas.core.series.Series) Expected classifier results.

Returns:

numpy.ndarray[bool]: Mask of selected features.

set_parameters(**kwargs)

Set the parameters/arguments of the algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

class niaaml.preprocessing.feature_selection.FeatureSelectionAlgorithm(**kwargs)

Bases: niaaml.preprocessing.preprocessing_algorithm.PreprocessingAlgorithm

Class for implementing feature selection algorithms.

Date:

2020

Author:

Luka Pečnik

License:

MIT

See Also:

• niaaml.preprocessing.preprocessing_algorithm.PreprocessingAlgorithm

select_features(x, y, **kwargs)

Perform the feature selection process.

Arguments:

x (pandas.core.frame.DataFrame): Array of original features. y (pandas.core.series.Series) Expected classifier results.

Returns:

numpy.ndarray[bool]: Mask of selected features.

class niaaml.preprocessing.feature_selection.FeatureSelectionAlgorithmFactory(**kwargs)

Bases: niaaml.utilities.Factory

Class with string mappings to feature selection algorithms.

Attributes:

_entities (Dict[str, FeatureSelectionAlgorithm]): Mapping from strings to feature selection algorithms.

See Also:

• niaaml.utilities.Factory

class niaaml.preprocessing.feature_selection.GreyWolfOptimizer(**kwargs)

Bases: niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Implementation of feature selection using GWO algorithm.

Date:

2020

Author:

Luka Pečnik

Reference:

The implementation is adapted according to the following article: D. Fister, I. Fister, T. Jagrič, I. Fister Jr., J. Brest. A novel self-adaptive differential evolution for feature selection using threshold mechanism . In: Proceedings of the 2018 IEEE Symposium on Computational Intelligence (SSCI 2018), pp. 17-24, 2018.

Reference URL:

http://iztok-jr-fister.eu/static/publications/236.pdf

License:

MIT

See Also:

• niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Name = 'Grey Wolf Optimizer'

select_features(x, y, **kwargs)

Perform the feature selection process.

Arguments:

x (pandas.core.frame.DataFrame): Array of original features. y (pandas.core.series.Series) Expected classifier results.

Returns:

numpy.ndarray[bool]: Mask of selected features.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

class niaaml.preprocessing.feature_selection.ParticleSwarmOptimization(**kwargs)

Bases: niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Implementation of feature selection using PSO algorithm.

Date:

2020

Author:

Luka Pečnik

Reference:

The implementation is adapted according to the following article: D. Fister, I. Fister, T. Jagrič, I. Fister Jr., J. Brest. A novel self-adaptive differential evolution for feature selection using threshold mechanism . In: Proceedings of the 2018 IEEE Symposium on Computational Intelligence (SSCI 2018), pp. 17-24, 2018.

Reference URL:

http://iztok-jr-fister.eu/static/publications/236.pdf

License:

MIT

See Also:

• niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Name = 'Particle Swarm Optimization'

select_features(x, y, **kwargs)

Perform the feature selection process.

Arguments:

x (pandas.core.frame.DataFrame): Array of original features. y (pandas.core.series.Series) Expected classifier results.

Returns:

numpy.ndarray[bool]: Mask of selected features.

set_parameters(**kwargs)

Set the parameters/arguments of the algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

class niaaml.preprocessing.feature_selection.SelectKBest(**kwargs)

Bases: niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Implementation of feature selection using selection of k best features according to used score function.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.SelectKBest.html

See Also:

• niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Name = 'Select K Best'

select_features(x, y, **kwargs)

Perform the feature selection process.

Arguments:

x (pandas.core.frame.DataFrame): Array of original features. y (pandas.core.series.Series) Expected classifier results.

Returns:

numpy.ndarray[bool]: Mask of selected features.

set_parameters(**kwargs)

Set the parameters/arguments of the algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

class niaaml.preprocessing.feature_selection.SelectPercentile(**kwargs)

Bases: niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Implementation of feature selection using percentile selection of best features according to used score function.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.SelectPercentile.html

See Also:

• niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Name = 'Select Percentile'

select_features(x, y, **kwargs)

Perform the feature selection process.

Arguments:

x (pandas.core.frame.DataFrame): Array of original features. y (pandas.core.series.Series) Expected classifier results.

Returns:

numpy.ndarray[bool]: Mask of selected features.

set_parameters(**kwargs)

Set the parameters/arguments of the algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

class niaaml.preprocessing.feature_selection.VarianceThreshold(**kwargs)

Bases: niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Implementation of feature selection using variance threshold.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.VarianceThreshold.html

See Also:

• niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Name = 'Variance Threshold'

select_features(x, y, **kwargs)

Perform the feature selection process.

Arguments:

x (pandas.core.frame.DataFrame): Array of original features. y (pandas.core.series.Series) Expected classifier results.

Returns:

numpy.ndarray[bool]: Mask of selected features.

set_parameters(**kwargs)

Set the parameters/arguments of the algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

class niaaml.preprocessing.feature_selection.jDEFSTH(**kwargs)

Bases: niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Implementation of self-adaptive differential evolution for feature selection using threshold mechanism.

Date:

2020

Author:

Iztok Fister Jr.

Reference:

4. Fister, I. Fister, T. Jagrič, I. Fister Jr., J. Brest. A novel self-adaptive differential evolution for feature selection using threshold mechanism . In: Proceedings of the 2018 IEEE Symposium on Computational Intelligence (SSCI 2018), pp. 17-24, 2018.

Reference URL:

http://iztok-jr-fister.eu/static/publications/236.pdf

License:

MIT

See Also:

• niaaml.preprocessing.feature_selection.feature_selection_algorithm.FeatureSelectionAlgorithm

Name = 'Self-Adaptive Differential Evolution'

select_features(x, y, **kwargs)

Perform the feature selection process.

Arguments:

x (pandas.core.frame.DataFrame): Array of original features. y (pandas.core.series.Series) Expected classifier results.

Returns:

numpy.ndarray[bool]: Mask of selected features.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

niaaml.preprocessing.feature_transform

class niaaml.preprocessing.feature_transform.FeatureTransformAlgorithm(**kwargs)

Bases: niaaml.preprocessing.preprocessing_algorithm.PreprocessingAlgorithm

Class for implementing feature transform algorithms.

Date:

2020

Author:

Luka Pečnik

License:

MIT

See Also:

• niaaml.preprocessing.preprocessing_algorithm.PreprocessingAlgorithm

fit(x, **kwargs)

Fit implemented feature transform algorithm.

Arguments:

x (pandas.core.frame.DataFrame): n samples to fit transformation algorithm.

transform(x, **kwargs)

Transforms the given x data.

Arguments:

x (pandas.core.frame.DataFrame): Data to transform.

Returns:

pandas.core.frame.DataFrame: Transformed data.

class niaaml.preprocessing.feature_transform.FeatureTransformAlgorithmFactory(**kwargs)

Bases: niaaml.utilities.Factory

Class with string mappings to feature transform algorithms.

Attributes:

_entities (Dict[str, FeatureTransformAlgorithm]): Mapping from strings to feature transform algorithms.

class niaaml.preprocessing.feature_transform.MaxAbsScaler(**kwargs)

Bases: niaaml.preprocessing.feature_transform.feature_transform_algorithm.FeatureTransformAlgorithm

Implementation of feature scaling by its maximum absolute value.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MaxAbsScaler.html#sklearn.preprocessing.MaxAbsScaler

See Also:

• niaaml.preprocessing.feature_transform.FeatureTransformAlgorithm

Name = 'Maximum Absolute Scaler'

fit(x, **kwargs)

Fit implemented transformation algorithm.

Arguments:

x (pandas.core.frame.DataFrame): n samples to fit transformation algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

transform(x, **kwargs)

Transforms the given x data.

Arguments:

x (pandas.core.frame.DataFrame): Data to transform.

Returns:

pandas.core.frame.DataFrame: Transformed data.

class niaaml.preprocessing.feature_transform.Normalizer(**kwargs)

Bases: niaaml.preprocessing.feature_transform.feature_transform_algorithm.FeatureTransformAlgorithm

Implementation of feature normalization algorithm.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.Normalizer

See Also:

• niaaml.preprocessing.feature_transform.FeatureTransformAlgorithm

Name = 'Normalizer'

fit(x, **kwargs)

Fit implemented transformation algorithm.

Arguments:

x (pandas.core.frame.DataFrame): n samples to fit transformation algorithm.

set_parameters(**kwargs)

Set the parameters/arguments of the algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

transform(x, **kwargs)

Transforms the given x data.

Arguments:

x (pandas.core.frame.DataFrame): Data to transform.

Returns:

pandas.core.frame.DataFrame: Transformed data.

class niaaml.preprocessing.feature_transform.QuantileTransformer(**kwargs)

Bases: niaaml.preprocessing.feature_transform.feature_transform_algorithm.FeatureTransformAlgorithm

Implementation of quantile transformer.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.QuantileTransformer.html#sklearn.preprocessing.QuantileTransformer

See Also:

• niaaml.preprocessing.feature_transform.FeatureTransformAlgorithm

Name = 'Quantile Transformer'

fit(x, **kwargs)

Fit implemented transformation algorithm.

Arguments:

x (pandas.core.frame.DataFrame): n samples to fit transformation algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

transform(x, **kwargs)

Transforms the given x data.

Arguments:

x (pandas.core.frame.DataFrame): Data to transform.

Returns:

pandas.core.frame.DataFrame: Transformed data.

class niaaml.preprocessing.feature_transform.RobustScaler(**kwargs)

Bases: niaaml.preprocessing.feature_transform.feature_transform_algorithm.FeatureTransformAlgorithm

Implementation of the robust scaler.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.RobustScaler.html#sklearn.preprocessing.RobustScaler

See Also:

• niaaml.preprocessing.feature_transform.FeatureTransformAlgorithm

Name = 'Robust Scaler'

fit(x, **kwargs)

Fit implemented transformation algorithm.

Arguments:

x (pandas.core.frame.DataFrame): n samples to fit transformation algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

transform(x, **kwargs)

Transforms the given x data.

Arguments:

x (pandas.core.frame.DataFrame): Data to transform.

Returns:

pandas.core.frame.DataFrame: Transformed data.

class niaaml.preprocessing.feature_transform.StandardScaler(**kwargs)

Bases: niaaml.preprocessing.feature_transform.feature_transform_algorithm.FeatureTransformAlgorithm

Implementation of feature standard scaling algorithm.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html

See Also:

• niaaml.preprocessing.feature_transform.FeatureTransformAlgorithm

Name = 'Standard Scaler'

fit(x, **kwargs)

Fit implemented transformation algorithm.

Arguments:

x (pandas.core.frame.DataFrame): n samples to fit transformation algorithm.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

transform(x, **kwargs)

Transforms the given x data.

Arguments:

x (pandas.core.frame.DataFrame): Data to transform.

Returns:

pandas.core.frame.DataFrame: Transformed data.

niaaml.preprocessing.encoding

class niaaml.preprocessing.encoding.EncoderFactory(**kwargs)

Bases: niaaml.utilities.Factory

Class with string mappings to encoders.

Attributes:

_entities (Dict[str, FeatureEncoder]): Mapping from strings to encoders.

See Also:

• niaaml.utilities.Factory

class niaaml.preprocessing.encoding.FeatureEncoder(**kwargs)

Bases: object

Class for implementing feature encoders.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Attributes:

Name (str): Name of the feature encoder.

Name = None

fit(feature)

Fit feature encoder.

Arguments:

feature (pandas.core.frame.DataFrame): A column (categorical) from DataFrame of features.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

transform(feature)

Transform feature’s values.

Arguments:

feature (pandas.core.frame.DataFrame): A column (categorical) from DataFrame of features.

Returns:

pandas.core.frame.DataFrame: A transformed column.

class niaaml.preprocessing.encoding.OneHotEncoder(**kwargs)

Bases: niaaml.preprocessing.encoding.feature_encoder.FeatureEncoder

Implementation of one-hot encoder.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Reference:

Seger, Cedric. “An investigation of categorical variable encoding techniques in machine learning: binary versus one-hot and feature hashing.” (2018).

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html

See Also:

• niaaml.preprocessing.encoding.FeatureEncoder

Name = 'One-Hot Encoder'

fit(feature)

Fit feature encoder.

Arguments:

feature (pandas.core.frame.DataFrame): A column (categorical) from DataFrame of features.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

transform(feature)

Transform feature’s values.

Arguments:

feature (pandas.core.frame.DataFrame): A column (categorical) from DataFrame of features.

Returns:

pandas.core.frame.DataFrame: A transformed column.

niaaml.preprocessing.encoding.encode_categorical_features(features, encoder)

Encode categorical features.

Arguments:

features (pandas.core.frame.DataFrame): DataFrame of features. encoder (str): Name of the encoder to use.

Returns:

Tuple[pandas.core.frame.DataFrame, Iterable[FeatureEncoder]]:

1. Converted dataframe.

2. Dictionary of encoders for all categorical features.

niaaml.preprocessing.imputation

class niaaml.preprocessing.imputation.Imputer(**kwargs)

Bases: object

Class for implementing imputers.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Attributes:

Name (str): Name of the imputer.

Name = None

fit(feature)

Fit imputer.

Arguments:

feature (pandas.core.frame.DataFrame): A column from DataFrame of features.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

transform(feature)

Transform feature’s values.

Arguments:

feature (pandas.core.frame.DataFrame): A column from DataFrame of features.

Returns:

pandas.core.frame.DataFrame: A transformed column.

class niaaml.preprocessing.imputation.ImputerFactory(**kwargs)

Bases: niaaml.utilities.Factory

Class with string mappings to imputers.

Attributes:

_entities (Dict[str, Imputer]): Mapping from strings to imputers.

See Also:

• niaaml.utilities.Factory

class niaaml.preprocessing.imputation.SimpleImputer(**kwargs)

Bases: niaaml.preprocessing.imputation.imputer.Imputer

Implementation of simple imputer.

Date:

2020

Author:

Luka Pečnik

License:

MIT

Documentation:

https://scikit-learn.org/stable/modules/generated/sklearn.impute.SimpleImputer.html

See Also:

• niaaml.preprocessing.imputation.Imputer

Name = 'Simple Imputer'

fit(feature)

Fit imputer.

Arguments:

feature (pandas.core.frame.DataFrame): A column from DataFrame of features.

to_string()

User friendly representation of the object.

Returns:

str: User friendly representation of the object.

transform(feature)

Transform feature’s values.

Arguments:

feature (pandas.core.frame.DataFrame): A column from DataFrame of features.

Returns:

pandas.core.frame.DataFrame: A transformed column.

niaaml.preprocessing.imputation.impute_features(features, imputer)

Impute features with missing data.

Arguments:

features (pandas.core.frame.DataFrame): DataFrame of features. imputer (str): Name of the imputer to use.

Returns:

Tuple[pandas.core.frame.DataFrame, Dict[Imputer]]:

1. Converted dataframe.

2. Dictionary of imputers for all features with missing data.