Scikit-Learn Naive Bayes
The following Scikit-Learn module types are Naive Bayes classifiers based on Bayes’ theorem.
Modules
BernoulliNB
Scikit-Learn module documentation: BernoulliNB
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Laplace smoothing parameter | --skl_alpha | Float ≥ 0 | None | alpha |
| Whether to learn class prior probabilities | --skl_fit_prior | True | False | None | fit_prior |
CategoricalNB
Scikit-Learn module documentation: CategoricalNB
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Laplace smoothing parameter | --skl_alpha | Float ≥ 0 | None | alpha |
| Whether to learn class prior probabilities | --skl_fit_prior | True | False | None | fit_prior |
ComplementNB
Scikit-Learn module documentation: ComplementNB
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Laplace smoothing parameter | --skl_alpha | Float ≥ 0 | None | alpha |
| Whether to learn class prior probabilities | --skl_fit_prior | True | False | None | fit_prior |
GaussianNB
Scikit-Learn module documentation: GaussianNB
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Portion of variance added for numerical stability | --skl_var_smoothing | Float > 0 | None | var_smoothing |
MultinomialNB
Scikit-Learn module documentation: MultinomialNB
Arguments
| Description | Zero2Neuro Argument | Possible Values | Default Value | Scikit-Learn Parameter |
|---|---|---|---|---|
| Laplace smoothing parameter | --skl_alpha | Float ≥ 0 | None | alpha |
| Whether to learn class prior probabilities | --skl_fit_prior | True | False | None | fit_prior |