Skip to content

getml.data.subroles

Subroles allow for more fine-granular control of how certain columns will be used by the pipeline.

A column can have no subrole, one subrole or several subroles.

Example 
# The Relboost feature learning algorithm will
# ignore this column.
my_data_frame.set_subroles(
    "my_column", getml.data.subroles.exclude.relboost)

# The Substring preprocessor will be applied to this column.
# But other preprocessors, feature learners or predictors
# are not excluded from using it as well.
my_data_frame.set_subroles(
    "ucc", getml.data.subroles.include.substring)

# Only the EmailDomain preprocessor will be applied
# to "emails". All other preprocessors, feature learners,
# feature selectors and predictors will ignore this column.
my_data_frame.set_subroles("emails", getml.data.subroles.only.email)

exclude

category_trimmer module-attribute 

category_trimmer: Final[ExcludeCategoryTrimmer] = (
    "exclude category trimmer"
)

The CategoryTrimmer preprocessor will ignore this column.

fastprop module-attribute 

fastprop: Final[ExcludeFastProp] = 'exclude fastprop'

FastProp will ignore this column.

feature_learners module-attribute 

feature_learners: Final[ExcludeFeatureLearners] = (
    "exclude feature learners"
)

All feature learners (feature_learning) will ignore this column.

imputation module-attribute 

imputation: Final[ExcludeImputation] = 'exclude imputation'

The Imputation preprocessor will ignore this column.

mapping module-attribute 

mapping: Final[ExcludeMapping] = 'exclude mapping'

The Mapping preprocessor will ignore this column.

multirel module-attribute 

multirel: Final[ExcludeMultirel] = 'exclude multirel'

Multirel will ignore this column.

predictors module-attribute 

predictors: Final[ExcludePredictors] = 'exclude predictors'

All predictors will ignore this column.

preprocessors module-attribute 

preprocessors: Final[ExcludePreprocessors] = (
    "exclude preprocessors"
)

All preprocessors will ignore this column.

relboost module-attribute 

relboost: Final[ExcludeRelboost] = 'exclude relboost'

Relboost will ignore this column.

relmt module-attribute 

relmt: Final[ExcludeRelMT] = 'exclude relmt'

RelMT will ignore this column.

seasonal module-attribute 

seasonal: Final[ExcludeSeasonal] = 'exclude seasonal'

The Seasonal preprocessor will ignore this column.

text_field_splitter module-attribute 

text_field_splitter: Final[ExcludeTextFieldSplitter] = (
    "exclude text field splitter"
)

The TextFieldSplitter will ignore this column.

include

email module-attribute 

email: Final[IncludeEmail] = 'include email'

A column with this subrole will be used for the EmailDomain preprocessor.

substring module-attribute 

substring: Final[IncludeSubstring] = 'include substring'

A column with this subrole will be used for the Substring preprocessor.

only

email module-attribute 

email: Final[OnlyEmail] = 'only email'

A column with this subrole will only be used for the EmailDomain preprocessor and nothing else. It will be ignored by all other preprocessors, feature learners and predictors.

substring module-attribute 

substring: Final[OnlySubstring] = 'only substring'

A column with this subrole will only be used for the Substring preprocessor and nothing else. It will be ignored by all other preprocessors, feature learners and predictors.

sets

exclude module-attribute 

exclude: FrozenSet[ExcludeLike] = frozenset(
    {
        category_trimmer,
        fastprop,
        feature_learners,
        imputation,
        mapping,
        multirel,
        predictors,
        preprocessors,
        relboost,
        relmt,
        seasonal,
        text_field_splitter,
    }
)

Set of subroles that exclude columns from certain operations.

include module-attribute 

include: FrozenSet[IncludeLike] = frozenset(
    {email, substring}
)

Set of subroles that explicitly include columns for certain operations.

only module-attribute 

only: FrozenSet[OnlyLike] = frozenset({email, substring})

Set of subroles that restrict the operations that can be performed on columns.

all_ module-attribute 

all_: FrozenSet[Subrole] = exclude | include | only

Set of all possible subroles.