Scikit-Learn Linear Models
The following Scikit-Learn module types are all linear models
Modules
- ARDRegression
- BayesianRidge
- ElasticNet
- GammaRegressor
- HuberRegressor
- Lars
- Lasso
- LassoLars
- LassoLarsIC
- LinearRegression
- LogisticRegression
- MultiTaskElasticNet
- MultiTaskLasso
- OrthogonalMatchingPursuit
- PassiveAggressiveRegressor
- Perceptron
- PoissonRegressor
- QuantileRegressor
- RANSACRegressor
- Ridge
- RidgeClassifier
- SGDClassifier
- SGDRegressor
- TheilSenRegressor
- TweedieRegressor
ARDRegression
Scikit-Learn module documentation: ARDRegression
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
BayesianRidge
Scikit-Learn module documentation: BayesianRidge
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
ElasticNet
Scikit-Learn module documentation: ElasticNet
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| L1 regularization strength | --L1_regularization | Float ≥ 0 | Required | alpha |
| L1 to L2 mixing ratio (0 = pure L2, 1 = pure L1) | --skl_l1_ratio | Float in [0, 1] | Required | l1_ratio |
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
GammaRegressor
Scikit-Learn module documentation: GammaRegressor
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| L2 regularization strength | --L2_regularization | Float ≥ 0 | None | alpha |
| Solver algorithm | --skl_solver | lbfgs | newton-cholesky | None | solver |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
HuberRegressor
Scikit-Learn module documentation: HuberRegressor
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| L2 regularization strength | --L2_regularization | Float ≥ 0 | None | alpha |
| Threshold distinguishing inliers from outliers | --skl_epsilon | Float > 1.0 | None | epsilon |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
Lars
Scikit-Learn module documentation: Lars
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Number of CPU cores to use for computation | --cpus_per_task | Integer ≥ 1 | None | n_jobs |
Lasso
Scikit-Learn module documentation: Lasso
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| L1 regularization strength | --L1_regularization | Float ≥ 0 | Required | alpha |
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
LassoLars
Scikit-Learn module documentation: LassoLars
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| L1 regularization strength | --L1_regularization | Float ≥ 0 | Required | alpha |
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
LassoLarsIC
Scikit-Learn module documentation: LassoLarsIC
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Model selection criterion | --skl_criterion | aic | bic | None | criterion |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
LinearRegression
Scikit-Learn module documentation: LinearRegression
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | Required | fit_intercept |
| Number of CPU cores to use for computation | --cpus_per_task | Integer ≥ 1 | None | n_jobs |
LogisticRegression
Scikit-Learn module documentation: LogisticRegression
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Number of CPU cores to use for computation | --cpus_per_task | Integer ≥ 1 | None | n_jobs |
| Inverse regularization strength | --skl_C | Float > 0 | None | C |
| Regularization penalty type | --skl_penalty | l1 | l2 | elasticnet | None | None | penalty |
| Solver algorithm | --skl_solver | lbfgs | liblinear | newton-cg | newton-cholesky | sag | saga | None | solver |
| L1 to L2 mixing ratio for elasticnet penalty | --skl_l1_ratio | Float in [0, 1] | None | l1_ratio |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
MultiTaskElasticNet
Scikit-Learn module documentation: MultiTaskElasticNet
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| L1 regularization strength | --L1_regularization | Float ≥ 0 | Required | alpha |
| L1 to L2 mixing ratio (0 = pure L2, 1 = pure L1) | --skl_l1_ratio | Float in [0, 1] | Required | l1_ratio |
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
MultiTaskLasso
Scikit-Learn module documentation: MultiTaskLasso
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| L1 regularization strength | --L1_regularization | Float ≥ 0 | Required | alpha |
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
OrthogonalMatchingPursuit
Scikit-Learn module documentation: OrthogonalMatchingPursuit
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Target number of non-zero coefficients | --skl_n_nonzero_coefs | Integer ≥ 1 | None | n_nonzero_coefs |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
PassiveAggressiveRegressor
Scikit-Learn module documentation: PassiveAggressiveRegressor
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Inverse regularization strength | --skl_C | Float > 0 | None | C |
| Maximum number of training passes | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
Perceptron
Scikit-Learn module documentation: Perceptron
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Number of CPU cores to use for computation | --cpus_per_task | Integer ≥ 1 | None | n_jobs |
| Initial learning rate | --skl_eta0 | Float > 0 | None | eta0 |
| Regularization penalty type | --skl_penalty | l1 | l2 | elasticnet | None | penalty |
| Maximum number of training passes | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
PoissonRegressor
Scikit-Learn module documentation: PoissonRegressor
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| L2 regularization strength | --L2_regularization | Float ≥ 0 | None | alpha |
| Solver algorithm | --skl_solver | lbfgs | newton-cholesky | None | solver |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
QuantileRegressor
Scikit-Learn module documentation: QuantileRegressor
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Quantile to estimate | --skl_quantile | Float in [0, 1] | None | quantile |
| L1 regularization strength | --L1_regularization | Float ≥ 0 | None | alpha |
| Solver algorithm | --skl_solver | highs | highs-ds | highs-ipm | interior-point | revised simplex | None | solver |
RANSACRegressor
Scikit-Learn module documentation: RANSACRegressor
Arguments
No Zero2Neuro arguments are currently mapped for this model; all parameters use Scikit-Learn defaults.
Ridge
Scikit-Learn module documentation: Ridge
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| L2 regularization strength | --L2_regularization | Float ≥ 0 | Required | alpha |
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Solver algorithm | --skl_solver | auto | svd | cholesky | lsqr | sparse_cg | sag | saga | lbfgs | None | solver |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
RidgeClassifier
Scikit-Learn module documentation: RidgeClassifier
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| L2 regularization strength | --L2_regularization | Float ≥ 0 | None | alpha |
| Solver algorithm | --skl_solver | auto | svd | cholesky | lsqr | sparse_cg | sag | saga | lbfgs | None | solver |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
SGDClassifier
Scikit-Learn module documentation: SGDClassifier
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Number of CPU cores to use for computation | --cpus_per_task | Integer ≥ 1 | None | n_jobs |
| Loss function | --skl_loss | hinge | log_loss | modified_huber | squared_hinge | perceptron | None | loss |
squared_error | huber | epsilon_insensitive | squared_epsilon_insensitive | ||||
| Regularization penalty type | --skl_penalty | l1 | l2 | elasticnet | None | penalty |
| L2 regularization strength | --L2_regularization | Float ≥ 0 | None | alpha |
| L1 to L2 mixing ratio for elasticnet penalty | --skl_l1_ratio | Float in [0, 1] | None | l1_ratio |
| Initial learning rate | --skl_eta0 | Float > 0 | None | eta0 |
| Learning rate schedule | --skl_learning_rate | constant | optimal | invscaling | adaptive | None | learning_rate |
| Maximum number of training passes | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
SGDRegressor
Scikit-Learn module documentation: SGDRegressor
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Loss function | --skl_loss | squared_error | huber | epsilon_insensitive | squared_epsilon_insensitive | None | loss |
| Regularization penalty type | --skl_penalty | l1 | l2 | elasticnet | None | penalty |
| L2 regularization strength | --L2_regularization | Float ≥ 0 | None | alpha |
| L1 to L2 mixing ratio for elasticnet penalty | --skl_l1_ratio | Float in [0, 1] | None | l1_ratio |
| Initial learning rate | --skl_eta0 | Float > 0 | None | eta0 |
| Learning rate schedule | --skl_learning_rate | constant | optimal | invscaling | adaptive | None | learning_rate |
| Maximum number of training passes | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
TheilSenRegressor
Scikit-Learn module documentation: TheilSenRegressor
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| Number of CPU cores to use for computation | --cpus_per_task | Integer ≥ 1 | None | n_jobs |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
TweedieRegressor
Scikit-Learn module documentation: TweedieRegressor
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit a bias/intercept term | --skl_include_bias | True | False | True | fit_intercept |
| L2 regularization strength | --L2_regularization | Float ≥ 0 | None | alpha |
| Tweedie distribution power parameter | --skl_tweedie_power | Float | None | power |
| Solver algorithm | --skl_solver | lbfgs | newton-cholesky | None | solver |
| Maximum number of solver iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |