Skip to main content

​
ForecastV2Response

class ForecastV2Response(BaseModel):
    forecasts: List[List[SingleSampleForecastPayload]]

​
Parameters

​
forecasts

Type: List[List[SingleSampleForecastPayload]] List of forecast scenarios, where each scenario contains forecasts for multiple time series samples. Each scenario corresponds to one forecasting scenario from the input request.

​
Notes

  • Each forecast row represents one forecasting scenario
  • All forecasts in a row should have the same timestamps
  • Empty forecasts (no timestamps/values) are represented as NaN columns
  • Forecasts can be converted back to DataFrames using to_dfs()

​
Examples

​
Basic Forecast Response

forecast1 = SingleSampleForecastPayload(
    sample_id="sales",
    timestamps=["2023-01-04", "2023-01-05"],
    values=[130.0, 140.0],
    model_name="sfm-moe-v1"
)
forecast2 = SingleSampleForecastPayload(
    sample_id="temperature",
    timestamps=["2023-01-04", "2023-01-05"],
    values=[23.0, 24.0],
    model_name="sfm-moe-v1"
)
response = ForecastV2Response(forecasts=[[forecast1, forecast2]])

​
Convert to DataFrames

dfs = response.to_dfs()

​
Methods

​
to_dfs

def to_dfs(self) -> List[pd.DataFrame]:
    """Convert the ForecastResponse to a list of DataFrames."""

​
See Also