Scikit-Learn Preprocessors
The following Scikit-Learn module types are feature preprocessors.
Modules
- MaxAbsScaler
- MinMaxScaler
- Normalizer
- PolynomialFeatures
- PowerTransformer
- QuantileTransformer
- RobustScaler
- SplineTransformer
- StandardScaler
MaxAbsScaler
Scikit-Learn module documentation: MaxAbsScaler
Arguments
No Zero2Neuro arguments are currently mapped for this model; all parameters use Scikit-Learn defaults.
MinMaxScaler
Scikit-Learn module documentation: MinMaxScaler
Arguments
No Zero2Neuro arguments are currently mapped for this model; all parameters use Scikit-Learn defaults.
Normalizer
Scikit-Learn module documentation: Normalizer
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Norm to use for normalization | --skl_norm | l1 | l2 | max | None | norm |
PolynomialFeatures
Scikit-Learn module documentation: PolynomialFeatures
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Degree of polynomial features | --skl_poly_degree | Integer ≥ 1 | Required | degree |
| Whether to produce only interaction features | --skl_poly_interaction_only | True | False | Required | interaction_only |
| Whether to include a bias (constant) column | --skl_include_bias | True | False | Required | include_bias |
PowerTransformer
Scikit-Learn module documentation: PowerTransformer
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Power transform method | --skl_method | yeo-johnson | box-cox | None | method |
| Whether to zero-mean, unit-variance the output | --skl_standardize | True | False | None | standardize |
QuantileTransformer
Scikit-Learn module documentation: QuantileTransformer
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Number of CPU cores to use | --cpus_per_task | Integer ≥ 1 | None | n_jobs |
| Number of quantiles for transformation | --skl_n_quantiles | Integer ≥ 1 | None | n_quantiles |
| Target output distribution | --skl_output_distribution | uniform | normal | None | output_distribution |
RobustScaler
Scikit-Learn module documentation: RobustScaler
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to center the data before scaling | --skl_with_centering | True | False | None | with_centering |
| Whether to scale the data to unit variance | --skl_with_scaling | True | False | None | with_scaling |
SplineTransformer
Scikit-Learn module documentation: SplineTransformer
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Whether to include a bias (constant) column | --skl_include_bias | True | False | True | include_bias |
| Number of knots | --skl_n_knots | Integer ≥ 2 | None | n_knots |
| Degree of the spline polynomial | --skl_poly_degree | Integer ≥ 0 | None | degree |
StandardScaler
Scikit-Learn module documentation: StandardScaler
Arguments
No Zero2Neuro arguments are currently mapped for this model; all parameters use Scikit-Learn defaults.