kumoai.pquery.PredictionTableGenerationPlan

class kumoai.pquery.PredictionTableGenerationPlan

Configuration parameters that define the construction of a Kumo prediction table from a predictive query.

Variables:

• anchor_time – (int | "inferred" | datetime.datetime) The time that a prediction horizon start time of “zero” refers to. If not set, will be inferred to be the latest timestamp in the target fact table. Note that this value can either be provided as an integer, representing the number of nanoseconds from the Unix epoch, or as a datetime.datetime object (default: "inferred"). Supported Task Types: Temporal

• forecast_length – (int) Turns a node regression problem into a forecasting problem (default: "missing"). Must be provided if the model was trained using this parameter. Supported Task Types: Temporal Regression

• lag_timesteps – (int) For forecasting problems, leverage the auto-regressive labels as inputs. This parameter controls the number of previous values that should be considered as auto-regressive labels (default: "missing"). Must be provided if the model was trained using this parameter. Supported Task Types: Temporal Regression

• year_over_year – (bool) For forecasting problems, integrate Year-Over-Year features as inputs to give more attention to the data from the previous year when making a prediction. Must be provided if the model was trained using this parameter. (default: "missing")