PrepareSeriesUnit
Turn a dataset into a series on its grid, and cut it into origins.
The task validates the roles, orders the rows by time and checks the
series sits on a regular grid (SingleSeriesForecastingTask.prepare_series).
What comes out is the series' values in time order, which is all a
forecaster reads, and the grid, which turns a step back into a date.
The splitter then cuts the series into backtest origins. The cuts are
steps, and they are published as split_indexes: the job stores them on
the run as it stores a holdout's rows, and the splitter's class is what
reads either.
The rows come out too, as dataset, in the same date order: the
exogenous variables are columns of those rows, and
ApplyCutPreprocessingUnit reads them from there, one step per row.
Nothing here applies the session's preprocessing. With preprocessing,
the job hands this unit only the time index as an input, since the
exogenous columns the model reads exist only once each cut's fit has run.
The sibling for the x/y tasks is PrepareAndSplitUnit. A series is not
an x and a y, and a backtest's cuts are not partitions of rows, so
this publishes different keys.
Parameters
- task_name : string, default=
TabularClassificationTask - Name of the task the dataset is prepared for.
- input_columns : array, default=
[] - Names of the columns used as model input.
- output_columns : array, default=
[] - Names of the columns the model has to predict.
- splitter : object
- Splitter that cuts the series into backtest origins, in steps.
Methods
execute(self, ctx: DashAI.back.units.context.ExecutionContext) -> None
PrepareSeriesUnitDo 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.