mojolearn.LinearSVR#

class mojolearn.LinearSVR(*, epsilon=0.0, penalty='l1', loss='epsilon_insensitive', C=1.0, fit_intercept=True, penalized_intercept=False, tol=0.0001, max_iter=1000, linesearch_max_iter=100, lbfgs_memory=5)[source]#

Bases: _LinearSVMBase

Linear support vector regression.

Minimizes the epsilon-insensitive or squared epsilon-insensitive loss plus an l1 or l2 penalty (see the mojolearn.svm module description for the objective). The intercept is never penalized.

The default penalty is 'l1' (cuML’s default), which uses OWL-QN and gives exact zeros. scikit-learn’s LinearSVR is l2 only: pass penalty='l2' for its objective.

X is read as float32; y must be one-dimensional. Invalid parameter values raise ValueError.

Parameters:

  • epsilon (float, default 0.0) – Width of the insensitive zone, finite and >= 0.

  • penalty ({‘l1’, ‘l2’}, default ‘l1’) – The regularization term.

  • loss ({‘epsilon_insensitive’, ‘squared_epsilon_insensitive’}, default ‘epsilon_insensitive’) – The data-fit loss.

  • C (float, default 1.0) – Inverse regularization strength, > 0.

  • fit_intercept (bool, default True) – Whether to fit an (unpenalized) intercept.

  • penalized_intercept (bool, default False) – Only False is supported; True raises NotImplementedError.

  • tol (float, default 1e-4) – Gradient tolerance, > 0.

  • max_iter (int, default 1000) – Maximum number of solver iterations.

  • linesearch_max_iter (int, default 100) – Maximum number of line-search steps per iteration.

  • lbfgs_memory (int, default 5) – Number of past updates kept by L-BFGS.

  • numeric_mode ({‘identical’, ‘fast’} or None, default None) – The numeric mode for this estimator; None uses the process default (see mojolearn.set_numeric_mode). 'fast' is for Apple GPUs.

Variables:

  • coef_ (array of shape (n_features,), float32) – The fitted coefficients.

  • intercept_ (float) – The fitted intercept; 0.0 when fit_intercept is False.

  • n_iter_ (array of shape (1,), int64) – The number of solver iterations run.

  • objective_ (float) – The final value of the minimized objective.

  • retcode_ (int) – The solver’s return code, as on LogisticRegression.

  • n_features_in_ (int) – The number of features seen in fit.

  • input_copied_ (bool) – Whether fit had to copy X to float32 C order.

Examples

>>> import numpy as np, mojolearn
>>> rng = np.random.default_rng(0)
>>> X = rng.random((1000, 8), dtype=np.float32)
>>> y = X @ np.arange(8, dtype=np.float32) + 1.0
>>> model = mojolearn.LinearSVR(penalty="l2", epsilon=0.1).fit(X, y)
>>> pred = model.predict(X)
__init__(*, epsilon=0.0, penalty='l1', loss='epsilon_insensitive', C=1.0, fit_intercept=True, penalized_intercept=False, tol=0.0001, max_iter=1000, linesearch_max_iter=100, lbfgs_memory=5)[source]#

Methods

fit(X, y[, sample_weight])

Fit the regressor.

get_params([deep])

numeric_mode_used()

The numeric mode this estimator runs in.

predict(X)

Predict with the fitted model.

score(X, y)

Return the coefficient of determination R^2 of the prediction.

set_params(**params)

Set constructor parameters.

vendor_used()

The GPU backend this estimator runs on.
fit(X, y, sample_weight=None)[source]#

Fit the regressor.

Parameters:

  • X (array-like of shape (n_samples, n_features)) – Training data, read as float32.

  • y (array-like of shape (n_samples,)) – Target values. A 2-D y or a length mismatch raises ValueError.

  • sample_weight (None) – Must be None; weights raise NotImplementedError.

Returns:

self (LinearSVR) – The fitted estimator.

predict(X)[source]#

Predict with the fitted model.

Parameters:

X (array-like of shape (n_samples, n_features)) – Samples, read as float32.

Returns:

y_pred (array of shape (n_samples,), float32) – X @ coef_ + intercept_.

Raises:

ValueError – If the model is not fitted, or X has a different number of features than the training data.

score(X, y)[source]#

Return the coefficient of determination R^2 of the prediction.

Parameters:

  • X (array-like of shape (n_samples, n_features)) – Test samples.

  • y (array-like of shape (n_samples,)) – True target values.

Returns:

score (float) – 1 - SS_res / SS_tot, computed in float64. A constant y scores 0.0.