InferDatasetTypesUnit
Publish a type declaration for every column of the dataset.
Prefers the types the dataset already carries: a dataloader that can read them from the source (a Parquet schema, a typed database column) knows better than any inference over the values, so re-inferring would throw that away. Only when there are none does it fall back to inferring from the data.
Publishes the declaration as a plain reference rather than applying it, so the decision and the transformation stay separable — the same declaration can be shown to a user for review before anything is cast.
Note the reference names columns, so it goes stale the moment something renames or drops one. It is meant to be consumed by the next step, not carried across a chain of transformations.
Parameters
- method : string, default=
DashAIPtype - Inference method used when the dataset does not already carry types of its own.
Methods
execute(self, ctx: DashAI.back.units.context.ExecutionContext) -> None
InferDatasetTypesUnitDo the unit's work, reading from and writing to the context.
Parameters
- ctx : ExecutionContext
- The shared execution context.
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(self, ctx: DashAI.back.units.context.ExecutionContext) -> None
BaseUnitCheck preconditions without executing the unit.
Parameters
- ctx : ExecutionContext
- The shared execution context.
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.