sklearn.linear_model.SGDOneClassSVM

class sklearn.linear_model.SGDOneClassSVM(nu=0.5, fit_intercept=True, max_iter=1000, tol=0.001, shuffle=True, verbose=0, random_state=None, learning_rate='optimal', eta0=0.0, power_t=0.5, warm_start=False, average=False)[source]

Solves linear One-Class SVM using Stochastic Gradient Descent.

This implementation is meant to be used with a kernel approximation technique (e.g. sklearn.kernel_approximation.Nystroem) to obtain results similar to sklearn.svm.OneClassSVM which uses a Gaussian kernel by default.

Read more in the User Guide.

New in version 1.0.

Parameters:
nufloat, default=0.5

The nu parameter of the One Class SVM: an upper bound on the fraction of training errors and a lower bound of the fraction of support vectors. Should be in the interval (0, 1]. By default 0.5 will be taken.

fit_interceptbool, default=True

Whether the intercept should be estimated or not. Defaults to True.

max_iterint, default=1000

The maximum number of passes over the training data (aka epochs). It only impacts the behavior in the fit method, and not the partial_fit. Defaults to 1000.

tolfloat or None, default=1e-3

The stopping criterion. If it is not None, the iterations will stop when (loss > previous_loss - tol). Defaults to 1e-3.

shufflebool, default=True

Whether or not the training data should be shuffled after each epoch. Defaults to True.

verboseint, default=0

The verbosity level.

random_stateint, RandomState instance or None, default=None

The seed of the pseudo random number generator to use when shuffling the data. If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by np.random.

learning_rate{‘constant’, ‘optimal’, ‘invscaling’, ‘adaptive’}, default=’optimal’

The learning rate schedule to use with fit. (If using partial_fit, learning rate must be controlled directly).

  • ‘constant’: eta = eta0

  • ‘optimal’: eta = 1.0 / (alpha * (t + t0)) where t0 is chosen by a heuristic proposed by Leon Bottou.

  • ‘invscaling’: eta = eta0 / pow(t, power_t)

  • ‘adaptive’: eta = eta0, as long as the training keeps decreasing. Each time n_iter_no_change consecutive epochs fail to decrease the training loss by tol or fail to increase validation score by tol if early_stopping is True, the current learning rate is divided by 5.

eta0float, default=0.0

The initial learning rate for the ‘constant’, ‘invscaling’ or ‘adaptive’ schedules. The default value is 0.0 as eta0 is not used by the default schedule ‘optimal’.

power_tfloat, default=0.5

The exponent for inverse scaling learning rate [default 0.5].

warm_startbool, default=False

When set to True, reuse the solution of the previous call to fit as initialization, otherwise, just erase the previous solution. See the Glossary.

Repeatedly calling fit or partial_fit when warm_start is True can result in a different solution than when calling fit a single time because of the way the data is shuffled. If a dynamic learning rate is used, the learning rate is adapted depending on the number of samples already seen. Calling fit resets this counter, while partial_fit will result in increasing the existing counter.

averagebool or int, default=False

When set to True, computes the averaged SGD weights and stores the result in the coef_ attribute. If set to an int greater than 1, averaging will begin once the total number of samples seen reaches average. So average=10 will begin averaging after seeing 10 samples.

Attributes:
coef_ndarray of shape (1, n_features)

Weights assigned to the features.

offset_ndarray of shape (1,)

Offset used to define the decision function from the raw scores. We have the relation: decision_function = score_samples - offset.

n_iter_int

The actual number of iterations to reach the stopping criterion.

t_int

Number of weight updates performed during training. Same as (n_iter_ * n_samples).

loss_function_concrete LossFunction
n_features_in_int

Number of features seen during fit.

New in version 0.24.

feature_names_in_ndarray of shape (n_features_in_,)

Names of features seen during fit. Defined only when X has feature names that are all strings.

New in version 1.0.

See also

sklearn.svm.OneClassSVM

Unsupervised Outlier Detection.

Notes

This estimator has a linear complexity in the number of training samples and is thus better suited than the sklearn.svm.OneClassSVM implementation for datasets with a large number of training samples (say > 10,000).

Examples

>>> import numpy as np
>>> from sklearn import linear_model
>>> X = np.array([[-1, -1], [-2, -1], [1, 1], [2, 1]])
>>> clf = linear_model.SGDOneClassSVM(random_state=42)
>>> clf.fit(X)
SGDOneClassSVM(random_state=42)
>>> print(clf.predict([[4, 4]]))
[1]

Methods

decision_function(X)

Signed distance to the separating hyperplane.

densify()

Convert coefficient matrix to dense array format.

fit(X[, y, coef_init, offset_init, ...])

Fit linear One-Class SVM with Stochastic Gradient Descent.

fit_predict(X[, y])

Perform fit on X and returns labels for X.

get_params([deep])

Get parameters for this estimator.

partial_fit(X[, y, sample_weight])

Fit linear One-Class SVM with Stochastic Gradient Descent.

predict(X)

Return labels (1 inlier, -1 outlier) of the samples.

score_samples(X)

Raw scoring function of the samples.

set_params(**params)

Set the parameters of this estimator.

sparsify()

Convert coefficient matrix to sparse format.

decision_function(X)[source]

Signed distance to the separating hyperplane.

Signed distance is positive for an inlier and negative for an outlier.

Parameters:
X{array-like, sparse matrix}, shape (n_samples, n_features)

Testing data.

Returns:
decarray-like, shape (n_samples,)

Decision function values of the samples.

densify()[source]

Convert coefficient matrix to dense array format.

Converts the coef_ member (back) to a numpy.ndarray. This is the default format of coef_ and is required for fitting, so calling this method is only required on models that have previously been sparsified; otherwise, it is a no-op.

Returns:
self

Fitted estimator.

fit(X, y=None, coef_init=None, offset_init=None, sample_weight=None)[source]

Fit linear One-Class SVM with Stochastic Gradient Descent.

This solves an equivalent optimization problem of the One-Class SVM primal optimization problem and returns a weight vector w and an offset rho such that the decision function is given by <w, x> - rho.

Parameters:
X{array-like, sparse matrix}, shape (n_samples, n_features)

Training data.

yIgnored

Not used, present for API consistency by convention.

coef_initarray, shape (n_classes, n_features)

The initial coefficients to warm-start the optimization.

offset_initarray, shape (n_classes,)

The initial offset to warm-start the optimization.

sample_weightarray-like, shape (n_samples,), optional

Weights applied to individual samples. If not provided, uniform weights are assumed. These weights will be multiplied with class_weight (passed through the constructor) if class_weight is specified.

Returns:
selfobject

Returns a fitted instance of self.

fit_predict(X, y=None)[source]

Perform fit on X and returns labels for X.

Returns -1 for outliers and 1 for inliers.

Parameters:
X{array-like, sparse matrix} of shape (n_samples, n_features)

The input samples.

yIgnored

Not used, present for API consistency by convention.

Returns:
yndarray of shape (n_samples,)

1 for inliers, -1 for outliers.

get_params(deep=True)[source]

Get parameters for this estimator.

Parameters:
deepbool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns:
paramsdict

Parameter names mapped to their values.

partial_fit(X, y=None, sample_weight=None)[source]

Fit linear One-Class SVM with Stochastic Gradient Descent.

Parameters:
X{array-like, sparse matrix}, shape (n_samples, n_features)

Subset of the training data.

yIgnored

Not used, present for API consistency by convention.

sample_weightarray-like, shape (n_samples,), optional

Weights applied to individual samples. If not provided, uniform weights are assumed.

Returns:
selfobject

Returns a fitted instance of self.

predict(X)[source]

Return labels (1 inlier, -1 outlier) of the samples.

Parameters:
X{array-like, sparse matrix}, shape (n_samples, n_features)

Testing data.

Returns:
yarray, shape (n_samples,)

Labels of the samples.

score_samples(X)[source]

Raw scoring function of the samples.

Parameters:
X{array-like, sparse matrix}, shape (n_samples, n_features)

Testing data.

Returns:
score_samplesarray-like, shape (n_samples,)

Unshiffted scoring function values of the samples.

set_params(**params)[source]

Set the parameters of this estimator.

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

Parameters:
**paramsdict

Estimator parameters.

Returns:
selfestimator instance

Estimator instance.

sparsify()[source]

Convert coefficient matrix to sparse format.

Converts the coef_ member to a scipy.sparse matrix, which for L1-regularized models can be much more memory- and storage-efficient than the usual numpy.ndarray representation.

The intercept_ member is not converted.

Returns:
self

Fitted estimator.

Notes

For non-sparse models, i.e. when there are not many zeros in coef_, this may actually increase memory usage, so use this method with care. A rule of thumb is that the number of zero elements, which can be computed with (coef_ == 0).sum(), must be more than 50% for this to provide significant benefits.

After calling this method, further fitting with the partial_fit method (if any) will not work until you call densify.

Examples using sklearn.linear_model.SGDOneClassSVM

One-Class SVM versus One-Class SVM using Stochastic Gradient Descent

One-Class SVM versus One-Class SVM using Stochastic Gradient Descent

One-Class SVM versus One-Class SVM using Stochastic Gradient Descent
Comparing anomaly detection algorithms for outlier detection on toy datasets

Comparing anomaly detection algorithms for outlier detection on toy datasets

Comparing anomaly detection algorithms for outlier detection on toy datasets