pyrcn.linear_model¶

The pyrcn.linear_model module has incremental linear regression.

class pyrcn.linear_model.IncrementalRegression(*, alpha: float = 1e-05, fit_intercept: bool = True, normalize: bool = False)¶

Bases: BaseEstimator, RegressorMixin

Linear regression.

This linear regression algorithm is able to perform a linear regression with the L2 regularization and iterative fit. [1] .. [1] https://ieeexplore.ieee.org/document/4012031

References

N. Liang, G. Huang, P. Saratchandran and N. Sundararajan, “A Fast and Accurate Online Sequential Learning Algorithm for Feedforward Networks,” in IEEE Transactions on Neural Networks, vol. 17, no. 6, pp. 1411-1423, Nov. 2006, doi: 10.1109/TNN.2006.880583.

Parameters

alpha (float, default=1e-5) – L2 regularization parameter
fit_intercept (bool, default=True) – Fits a constant offset if True. Use this if input values are not average free.
normalize (bool, default=False) – Performs a preprocessing normalization if True.

coef_¶

Weight vector(s).

Type: array, shape (n_features,) or (n_targets, n_features)

intercept_¶

Independent term in decision function. Set to 0.0 if fit_intercept = False.

Type: float | array, shape = (n_targets,)

property coef_: Optional[ndarray]¶

Return the output weights without intercept.

Compatibility to `sklearn.linear_model.Ridge`.

Returns: coef_ – Weight vector(s).
Return type: ndarray of shape (n_features,) or (n_targets, n_features)

fit(X: ndarray, y: ndarray) → IncrementalRegression¶

Fit the regressor.

Parameters

X (ndarray of shape (samples, n_features)) –
y (ndarray of shape (n_samples,) or (n_samples, n_targets)) – The targets to predict.
partial_normalize (bool, default=True) – Partial fits the normalization transformer on this sample if True.
reset (bool, default=False) – Begin a new fit, drop prior fits.
validate (bool, default=True) – Validate input data if True.
postpone_inverse (bool, default=False) – If the output weights have not been fitted yet, regressor might be hinted at postponing inverse calculation.

Returns

self

Return type

returns a fitted IncrementalRegression model

property intercept_: ndarray¶

Return the intercept of output output weights.

Compatibility to `sklearn.linear_model.Ridge`.

Returns: intercept_ – Independent term in decision function.
Return type: Union[float, np.ndarray]

partial_fit(X: ndarray, y: ndarray, partial_normalize: bool = True, reset: bool = False, validate: bool = True, postpone_inverse: bool = False) → IncrementalRegression¶

Fit the regressor partially.

Parameters

X (ndarray of shape (samples, n_features)) –
y (ndarray of shape (n_samples,) or (n_samples, n_targets)) – The targets to predict.
partial_normalize (bool, default=True) – Partial fits the normalization transformer on this sample if True.
reset (bool, default=False) – Begin a new fit, drop prior fits.
validate (bool, default=True) – Validate input data if True.
postpone_inverse (bool, default=False) – If the output weights have not been fitted yet, regressor might be hinted at postponing inverse calculation.

Returns

self

Return type

returns a partially fitted IncrementalRegression model

predict(X: ndarray) → ndarray¶

Predict output y according to input X.

Parameters: X (ndarray of shape (samples, n_features)) –
Returns: y
Return type: ndarray of shape (n_samples,) or (n_samples, n_targets)

set_partial_fit_request(*, partial_normalize: Union[bool, None, str] = '$UNCHANGED$', postpone_inverse: Union[bool, None, str] = '$UNCHANGED$', reset: Union[bool, None, str] = '$UNCHANGED$', validate: Union[bool, None, str] = '$UNCHANGED$') → IncrementalRegression¶

Request metadata passed to the partial_fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

True: metadata is requested, and passed to partial_fit if provided. The request is ignored if metadata is not provided.
False: metadata is not requested and the meta-estimator will not pass it to partial_fit.
None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.
str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters

partial_normalize (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for partial_normalize parameter in partial_fit.
postpone_inverse (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for postpone_inverse parameter in partial_fit.
reset (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for reset parameter in partial_fit.
validate (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for validate parameter in partial_fit.

Returns

self – The updated object.

Return type

object

set_score_request(*, sample_weight: Union[bool, None, str] = '$UNCHANGED$') → IncrementalRegression¶

Request metadata passed to the score method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

True: metadata is requested, and passed to score if provided. The request is ignored if metadata is not provided.
False: metadata is not requested and the meta-estimator will not pass it to score.
None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.
str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters: sample_weight (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_weight parameter in score.
Returns: self – The updated object.
Return type: object