For getting forecasts with the data platform backend, the default is to pull the forecasts from the forecaster that most recently ran. However, now we are updating the versions of the forecasters less regularly, we should remove dependency on this. Ideally, the forecaster name and version desired for the location type/region etc should be mapped or hardcoded in the routers.
That way, this part could be removed, and the forecaster_name and forecaster_version parameters could be made non-optional.
https://github.com/openclimatefix/quartz-api/blob/main/src/quartz_api/internal/backends/dataplatform/client.py#L102-L119
this will speed up queries.
For getting forecasts with the data platform backend, the default is to pull the forecasts from the forecaster that most recently ran. However, now we are updating the versions of the forecasters less regularly, we should remove dependency on this. Ideally, the forecaster name and version desired for the location type/region etc should be mapped or hardcoded in the routers.
That way, this part could be removed, and the forecaster_name and forecaster_version parameters could be made non-optional.
https://github.com/openclimatefix/quartz-api/blob/main/src/quartz_api/internal/backends/dataplatform/client.py#L102-L119
this will speed up queries.