TransformDatasetUnit
Apply an already fitted converter to the dataset in the context.
Takes no converter configuration: the converter arrives fitted through the
context, from :class:FitConverterUnit or :class:ApplyConverterUnit. That
is what makes "fit on train, apply to test" possible without refitting —
run this once per dataset, and the converter keeps the state it learned.
It does carry its own scope, on purpose: the scope is a list of 1-based
indexes, and they are resolved against this unit's dataset. Reusing the
column names resolved during fit would break the moment the two datasets
order their columns differently, and would carry stale column identity
across the context boundary.
Parameters
- 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
TransformDatasetUnitDo 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
TransformDatasetUnitReject 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.