FitConverterUnit
Fit a converter on the dataset in the context and publish it fitted.
Splits the "fit" half out of :class:ApplyConverterUnit so a converter can
be learned from one dataset and applied to another. That is the standard
train/test discipline: a scaler, an encoder or an imputer must learn its
statistics from the training data only, and then be applied unchanged to the
test data — refitting on test would leak information into the evaluation.
The dataset is left untouched: this unit produces a fitted converter, not
data. Pair it with :class:TransformDatasetUnit, once per dataset the
fitted converter has to be applied to.
Parameters
- converter : object
- Converter to apply, together with its own configuration.
- scope
- Part of the dataset the converter applies to: 'columns' is a list of {'idx': n} and 'rows' a list of n, both 1-based. An empty column list means every column.
- target, default=
None - Target column handed to the converter as y, as {'idx': n} with n 1-based. Leave empty for unsupervised converters.
Methods
execute(self, ctx: DashAI.back.units.context.ExecutionContext) -> None
FitConverterUnitDo the unit's work, reading from and writing to the context.
Parameters
- ctx : ExecutionContext
- The shared execution context.
validate(self, ctx: DashAI.back.units.context.ExecutionContext) -> None
FitConverterUnitReject an out-of-bounds target before any work is done.
get_credential(self, name: str)
ConfigObjectResolve a registered credential component by name.
Parameters
- name : str
- Credential component class name (e.g. "HuggingFaceCredential").
Returns
- BaseCredential
- An instance of the requested credential component.
get_schema(cls) -> dict
ConfigObjectGenerates the component related Json Schema.
Returns
- dict
- Dictionary representing the Json Schema of the component.
validate_and_transform(self, raw_data: dict) -> dict
ConfigObjectIt takes the data given by the user to initialize the model and returns it with all the objects that the model needs to work.
Parameters
- raw_data : dict
- A dictionary with the data provided by the user to initialize the model.
Returns
- dict
- A validated dictionary with the necessary objects.