Scikit-Learn Support Vector Machines
The following Scikit-Learn module types are support vector machine models.
Modules
LinearSVC
Scikit-Learn module documentation: LinearSVC
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit the intercept | --skl_include_bias | True | False | True | fit_intercept |
| Regularization parameter | --skl_C | Float > 0 | None | C |
| Norm used in penalization | --skl_penalty | l1 | l2 | None | penalty |
| Loss function | --skl_loss | hinge | squared_hinge | None | loss |
| Maximum number of iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
| Random seed for reproducibility | --skl_random_state | Integer | None | random_state |
LinearSVR
Scikit-Learn module documentation: LinearSVR
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to fit the intercept | --skl_include_bias | True | False | True | fit_intercept |
| Regularization parameter | --skl_C | Float > 0 | None | C |
| Loss function | --skl_loss | epsilon_insensitive | squared_epsilon_insensitive | None | loss |
| Epsilon in the epsilon-insensitive loss tube | --skl_epsilon | Float ≥ 0 | None | epsilon |
| Maximum number of iterations | --skl_max_iter | Integer ≥ 1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
| Random seed for reproducibility | --skl_random_state | Integer | None | random_state |
NuSVC
Scikit-Learn module documentation: NuSVC
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Upper bound on training errors and support vectors | --skl_nu | Float in (0, 1] | None | nu |
| Kernel function | --skl_kernel | linear | poly | rbf | sigmoid | precomputed | None | kernel |
| Degree of polynomial kernel | --skl_poly_degree | Integer ≥ 1 | None | degree |
| Kernel coefficient | --skl_gamma | scale | auto | numeric string | None | gamma |
| Independent term in kernel function | --skl_coef0 | Float | None | coef0 |
| Maximum number of iterations (-1 for no limit) | --skl_max_iter | Integer ≥ 1 or -1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
NuSVR
Scikit-Learn module documentation: NuSVR
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Fraction of training errors and support vectors | --skl_nu | Float in (0, 1] | None | nu |
| Regularization parameter | --skl_C | Float > 0 | None | C |
| Kernel function | --skl_kernel | linear | poly | rbf | sigmoid | precomputed | None | kernel |
| Degree of polynomial kernel | --skl_poly_degree | Integer ≥ 1 | None | degree |
| Kernel coefficient | --skl_gamma | scale | auto | numeric string | None | gamma |
| Independent term in kernel function | --skl_coef0 | Float | None | coef0 |
| Maximum number of iterations (-1 for no limit) | --skl_max_iter | Integer ≥ 1 or -1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
OneClassSVM
Scikit-Learn module documentation: OneClassSVM
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Upper bound on training errors | --skl_nu | Float in (0, 1] | None | nu |
| Kernel function | --skl_kernel | linear | poly | rbf | sigmoid | precomputed | None | kernel |
| Degree of polynomial kernel | --skl_poly_degree | Integer ≥ 1 | None | degree |
| Kernel coefficient | --skl_gamma | scale | auto | numeric string | None | gamma |
| Independent term in kernel function | --skl_coef0 | Float | None | coef0 |
| Maximum number of iterations (-1 for no limit) | --skl_max_iter | Integer ≥ 1 or -1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
SVC
Scikit-Learn module documentation: SVC
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Regularization parameter | --skl_C | Float > 0 | None | C |
| Kernel function | --skl_kernel | linear | poly | rbf | sigmoid | precomputed | None | kernel |
| Degree of polynomial kernel | --skl_poly_degree | Integer ≥ 1 | None | degree |
| Kernel coefficient | --skl_gamma | scale | auto | numeric string | None | gamma |
| Independent term in kernel function | --skl_coef0 | Float | None | coef0 |
| Maximum number of iterations (-1 for no limit) | --skl_max_iter | Integer ≥ 1 or -1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |
SVR
Scikit-Learn module documentation: SVR
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Regularization parameter | --skl_C | Float > 0 | None | C |
| Kernel function | --skl_kernel | linear | poly | rbf | sigmoid | precomputed | None | kernel |
| Degree of polynomial kernel | --skl_poly_degree | Integer ≥ 1 | None | degree |
| Kernel coefficient | --skl_gamma | scale | auto | numeric string | None | gamma |
| Independent term in kernel function | --skl_coef0 | Float | None | coef0 |
| Epsilon in the epsilon-insensitive loss tube | --skl_epsilon | Float ≥ 0 | None | epsilon |
| Maximum number of iterations (-1 for no limit) | --skl_max_iter | Integer ≥ 1 or -1 | None | max_iter |
| Convergence tolerance | --skl_tol | Float > 0 | None | tol |