Current problem
Right now, we are required to provide values for dataset_y in both initialize_workflow AND update_workflow_with_data (and the batch function) in the instance that x values are provided. (Y values can only be left off if the X values are also left off.) With certain workflows, we may not necessarily require the y values but would like to call functions such as get_next_point, get_surrogate_values, etc. using the X values without having to provide Y values.
Changes
These changes will have to take place in the DIAL Dataclass API, and will also require us to tweak some logic in the core functionality which may require equal lengths of dataset_x and dataset_y.
Current problem
Right now, we are required to provide values for
dataset_yin bothinitialize_workflowANDupdate_workflow_with_data(and the batch function) in the instance that x values are provided. (Y values can only be left off if the X values are also left off.) With certain workflows, we may not necessarily require the y values but would like to call functions such asget_next_point,get_surrogate_values, etc. using the X values without having to provide Y values.Changes
These changes will have to take place in the DIAL Dataclass API, and will also require us to tweak some logic in the core functionality which may require equal lengths of dataset_x and dataset_y.