bace API¶

`ComplementNB`([alpha, weight_normalized])	Complement Naive Bayes classifier
`NegationNB`([alpha])	Negation Naive Bayes classifier
`UniversalSetNB`([alpha])	Universal-set Naive Bayes classifier
`SelectiveNB`([alpha])	Selective Naive Bayes classifier

class bace.ComplementNB(alpha=1.0, weight_normalized=False)[source]¶

Bases: bace.base.BaseNB

Complement Naive Bayes classifier

References

Rennie J. D. M., Shih L., Teevan J., Karger D. R. (2003). Tackling the Poor Assumptions of Naive Bayes Text Classifiers

https://people.csail.mit.edu/jrennie/papers/icml03-nb.pdf

Parameters:	alpha (float) – Smoothing parameter weight_normalized (bool, default False) – Enable Weight-normalized Complement Naive Bayes method.

alpha_sum_¶

Sum of alpha params

Type:	int

classes_¶

Classes list

Type:	array, shape (n_classes,)

class_count_¶

number of training samples observed in each class.

Type:	array, shape (n_classes,)

Examples

>>> from sklearn.datasets import fetch_20newsgroups
>>> from sklearn.feature_extraction.text import CountVectorizer
>>> from bace import ComplementNB
Prepare data
>>> vectorizer = CountVectorizer()
>>> categories = ['alt.atheism', 'talk.religion.misc','comp.graphics', 'sci.space']
Train set
>>> newsgroups_train = fetch_20newsgroups(subset='train', categories=categories, shuffle=True)
>>> train_vectors = vectorizer.fit_transform(newsgroups_train.data)
Test set
>>> newsgroups_test = fetch_20newsgroups(subset='test', categories=categories, shuffle=True)
>>> test_vectors = vectorizer.transform(newsgroups_test.data)
>>> clf = ComplementNB()
>>> clf.fit(newsgroups_train, train_vectors).accuracy_score(newsgroups_test, test_vectors)

accuracy_score(X, y)¶

Return acuracy score

Parameters:	X ({array-like, sparse matrix}, shape = [n_samples, n_features]) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape = [n_samples]) – Target values.
Returns:	accuracy_score – Accuracy on the given test set
Return type:	float

class_log_proba_¶: Log probability of class occurrence

complement_class_count_¶: Complement class count, i.e. number of occurrences of all the samples with all the classes except the given class c

complement_class_log_proba_¶: Complement class probability, i.e. logprob of occurrence of a sample, which does not belong to the given class c

fit(X, y)¶

Fit model to given training set

Parameters:	X (array-like, shape (n_samples, n_features)) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape (n_samples,)) – Target values.
Returns:	self – Returns self.
Return type:	Naive Bayes estimator object

get_params(deep=True)¶

Get parameters for this estimator.

Parameters:	deep (boolean, optional) – If True, will return the parameters for this estimator and contained subobjects that are estimators.
Returns:	params – Parameter names mapped to their values.
Return type:	mapping of string to any

partial_fit(X, y, classes=None)¶

Incremental fit on a batch of samples.

Parameters:	X ({array-like, sparse matrix}, shape = [n_samples, n_features]) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape = [n_samples]) – Target values. classes (array-like, shape = [n_classes], optional (default=None)) – List of all the classes that can possibly appear in the y vector. Must be provided at the first call to partial_fit, can be omitted in subsequent calls.
Returns:	self – Returns self.
Return type:	object

predict(X)[source]¶

Perform classification on an array of test vectors X.

Parameters:	X (array-like, shape = [n_samples, n_features]) – Unseen samples vector
Returns:	C – Predicted target values for X
Return type:	array, shape = [n_samples]

predict_log_proba(X)[source]¶

Return log-probability estimates for the test vector X.

Parameters:	X (array-like, shape = [n_samples, n_features]) –
Returns:	C – Returns the log-probability of the samples for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute classes_.
Return type:	array-like, shape = [n_samples, n_classes]

predict_proba(X)¶

Return probability estimates for the test vector X. :param X: :type X: array-like, shape = [n_samples, n_features]

Returns:	C – Returns the probability of the samples for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute classes_.
Return type:	array-like, shape = [n_samples, n_classes]

set_params(**params)¶

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Returns:
Return type:	self

class bace.NegationNB(alpha=1.0)[source]¶

Bases: bace.base.BaseNB

Negation Naive Bayes classifier

Parameters:	alpha (float) – Smoothing parameter

References

Komiya K., Sato N., Fujimoto K., Kotani Y. (2011). Negation Naive Bayes for Categorization of Product Pages on the Web

http://www.aclweb.org/anthology/R11-1083.pdf

accuracy_score(X, y)¶

Return acuracy score

Parameters:	X ({array-like, sparse matrix}, shape = [n_samples, n_features]) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape = [n_samples]) – Target values.
Returns:	accuracy_score – Accuracy on the given test set
Return type:	float

class_log_proba_¶: Log probability of class occurrence

complement_class_count_¶: Complement class count, i.e. number of occurrences of all the samples with all the classes except the given class c

complement_class_log_proba_¶: Complement class probability, i.e. logprob of occurrence of a sample, which does not belong to the given class c

fit(X, y)¶

Fit model to given training set

Parameters:	X (array-like, shape (n_samples, n_features)) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape (n_samples,)) – Target values.
Returns:	self – Returns self.
Return type:	Naive Bayes estimator object

get_params(deep=True)¶

Get parameters for this estimator.

Parameters:	deep (boolean, optional) – If True, will return the parameters for this estimator and contained subobjects that are estimators.
Returns:	params – Parameter names mapped to their values.
Return type:	mapping of string to any

partial_fit(X, y, classes=None)¶

Incremental fit on a batch of samples.

Parameters:	X ({array-like, sparse matrix}, shape = [n_samples, n_features]) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape = [n_samples]) – Target values. classes (array-like, shape = [n_classes], optional (default=None)) – List of all the classes that can possibly appear in the y vector. Must be provided at the first call to partial_fit, can be omitted in subsequent calls.
Returns:	self – Returns self.
Return type:	object

predict(X)[source]¶

Perform classification on an array of test vectors X.

Parameters:	X (array-like, shape = [n_samples, n_features]) – Unseen samples vector
Returns:	C – Predicted target values for X
Return type:	array, shape = [n_samples]

predict_log_proba(X)[source]¶

Return log-probability estimates for the test vector X.

Parameters:	X (array-like, shape = [n_samples, n_features]) –
Returns:	C – Returns the log-probability of the samples for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute classes_.
Return type:	array-like, shape = [n_samples, n_classes]

predict_proba(X)¶

Return probability estimates for the test vector X. :param X: :type X: array-like, shape = [n_samples, n_features]

Returns:	C – Returns the probability of the samples for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute classes_.
Return type:	array-like, shape = [n_samples, n_classes]

set_params(**params)¶

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Returns:
Return type:	self

class bace.UniversalSetNB(alpha=1.0)[source]¶

Bases: bace.base.BaseNB

Universal-set Naive Bayes classifier

Parameters:	alpha (float) – Smoothing parameter

References

Komiya K., Ito Y., Kotani Y. (2013). New Naive Bayes Methods using Data from All Classes

https://github.com/krzjoa/bace/blob/master/papers/snb.pdf

accuracy_score(X, y)¶

Return acuracy score

Parameters:	X ({array-like, sparse matrix}, shape = [n_samples, n_features]) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape = [n_samples]) – Target values.
Returns:	accuracy_score – Accuracy on the given test set
Return type:	float

class_log_proba_¶: Log probability of class occurrence

complement_class_count_¶: Complement class count, i.e. number of occurrences of all the samples with all the classes except the given class c

complement_class_log_proba_¶: Complement class probability, i.e. logprob of occurrence of a sample, which does not belong to the given class c

fit(X, y)[source]¶

Fit model to given training set

Parameters:	X (array-like, shape (n_samples, n_features)) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape (n_samples,)) – Target values.
Returns:	self – Returns self.
Return type:	Naive Bayes estimator object

get_params(deep=True)¶

Get parameters for this estimator.

Parameters:	deep (boolean, optional) – If True, will return the parameters for this estimator and contained subobjects that are estimators.
Returns:	params – Parameter names mapped to their values.
Return type:	mapping of string to any

partial_fit(X, y, classes=None)[source]¶

Incremental fit on a batch of samples.

Parameters:	X ({array-like, sparse matrix}, shape = [n_samples, n_features]) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape = [n_samples]) – Target values. classes (array-like, shape = [n_classes], optional (default=None)) – List of all the classes that can possibly appear in the y vector. Must be provided at the first call to partial_fit, can be omitted in subsequent calls.
Returns:	self – Returns self.
Return type:	object

predict(X)[source]¶

Perform classification on an array of test vectors X.

Parameters:	X (array-like, shape = [n_samples, n_features]) – Unseen samples vector
Returns:	C – Predicted target values for X
Return type:	array, shape = [n_samples]

predict_log_proba(X)[source]¶

Return log-probability estimates for the test vector X.

Parameters:	X (array-like, shape = [n_samples, n_features]) –
Returns:	C – Returns the log-probability of the samples for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute classes_.
Return type:	array-like, shape = [n_samples, n_classes]

predict_proba(X)¶

Return probability estimates for the test vector X. :param X: :type X: array-like, shape = [n_samples, n_features]

Returns:	C – Returns the probability of the samples for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute classes_.
Return type:	array-like, shape = [n_samples, n_classes]

set_params(**params)¶

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Returns:
Return type:	self

class bace.SelectiveNB(alpha=1.0)[source]¶

Bases: bace.base.BaseNB

Selective Naive Bayes classifier

Parameters:	alpha (float) – Smoothing parameter

References

Komiya K., Ito Y., Kotani Y. (2013). New Naive Bayes Methods using Data from All Classes

https://github.com/krzjoa/bace/blob/master/papers/snb.pdf

accuracy_score(X, y)¶

Return acuracy score

Parameters:	X ({array-like, sparse matrix}, shape = [n_samples, n_features]) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape = [n_samples]) – Target values.
Returns:	accuracy_score – Accuracy on the given test set
Return type:	float

class_log_proba_¶: Log probability of class occurrence

complement_class_count_¶: Complement class count, i.e. number of occurrences of all the samples with all the classes except the given class c

complement_class_log_proba_¶: Complement class probability, i.e. logprob of occurrence of a sample, which does not belong to the given class c

fit(X, y)¶

Fit model to given training set

Parameters:	X (array-like, shape (n_samples, n_features)) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape (n_samples,)) – Target values.
Returns:	self – Returns self.
Return type:	Naive Bayes estimator object

get_params(deep=True)¶

Get parameters for this estimator.

Parameters:	deep (boolean, optional) – If True, will return the parameters for this estimator and contained subobjects that are estimators.
Returns:	params – Parameter names mapped to their values.
Return type:	mapping of string to any

partial_fit(X, y, classes=None)¶

Incremental fit on a batch of samples.

Parameters:	X ({array-like, sparse matrix}, shape = [n_samples, n_features]) – Training vectors, where n_samples is the number of samples and n_features is the number of features. y (array-like, shape = [n_samples]) – Target values. classes (array-like, shape = [n_classes], optional (default=None)) – List of all the classes that can possibly appear in the y vector. Must be provided at the first call to partial_fit, can be omitted in subsequent calls.
Returns:	self – Returns self.
Return type:	object

predict(X)[source]¶

Perform classification on an array of test vectors X.

Parameters:	X (array-like, shape = [n_samples, n_features]) – Unseen samples vector
Returns:	C – Predicted target values for X
Return type:	array, shape = [n_samples]

predict_log_proba(X)[source]¶

Return log-probability estimates for the test vector X.

Parameters:	X (array-like, shape = [n_samples, n_features]) –
Returns:	C – Returns the log-probability of the samples for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute classes_.
Return type:	array-like, shape = [n_samples, n_classes]

predict_proba(X)¶

Return probability estimates for the test vector X. :param X: :type X: array-like, shape = [n_samples, n_features]

Returns:	C – Returns the probability of the samples for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute classes_.
Return type:	array-like, shape = [n_samples, n_classes]

set_params(**params)¶

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Returns:
Return type:	self

class bace.Benchmark(classifiers, verbose=False)[source]¶

Bases: sklearn.base.BaseEstimator

scikit-learn like classifiers benchmark

Parameters:	classifiers (list of sklearn.base.BaseEstimator) – List of sklearn classifiers verbose (bool) – Print training details

compare(X, y, metrics={'Accuracy': <function accuracy_score>})[source]¶

Compare predictions of multiple classifiers

Parameters:	X (numpy.ndarray) – Features y (numpy.ndarray) – Targets metrics (dict of callable) – List of metric functions

fit(X, y)[source]¶

Fit several classifiers

Parameters:	X (numpy.ndarray) – y (numpy.ndarray) – Labels

get_params(deep=True)¶

Get parameters for this estimator.

Parameters:	deep (boolean, optional) – If True, will return the parameters for this estimator and contained subobjects that are estimators.
Returns:	params – Parameter names mapped to their values.
Return type:	mapping of string to any

predict(X)[source]¶

set_params(**params)¶

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Returns:
Return type:	self

class bace.BenchmarkNaiveBayes[source]¶

Bases: bace.benchmark.Benchmark

CLASSIFIERS = [MultinomialNB(alpha=1.0, class_prior=None, fit_prior=True), ComplementNB(alpha=1.0, weight_normalized=True), ComplementNB(alpha=1.0, weight_normalized=False), ComplementNB(alpha=1.0, class_prior=None, fit_prior=True, norm=False), ComplementNB(alpha=1.0, class_prior=None, fit_prior=True, norm=True), NegationNB(alpha=1.0), SelectiveNB(alpha=1.0), UniversalSetNB(alpha=1.0)]¶

compare(X, y, metrics={'Accuracy': <function accuracy_score>})¶

Compare predictions of multiple classifiers

Parameters:	X (numpy.ndarray) – Features y (numpy.ndarray) – Targets metrics (dict of callable) – List of metric functions

fit(X, y)¶

Fit several classifiers

Parameters:	X (numpy.ndarray) – y (numpy.ndarray) – Labels

get_params(deep=True)¶

Get parameters for this estimator.

Parameters:	deep (boolean, optional) – If True, will return the parameters for this estimator and contained subobjects that are estimators.
Returns:	params – Parameter names mapped to their values.
Return type:	mapping of string to any

predict(X)¶

set_params(**params)¶

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Returns:
Return type:	self