Is there a work around to getting the results from a dagstermill asset running in the pipeline? I am running into this:

def yield_result(self, value, output_name="result"):
 
    <snip>
 
    # dagstermill assets do not support yielding additional results within the notebook:
    if len(step_context.job_def.asset_layer.asset_keys) > 0:
      raise DagstermillError(
        "dagstermill assets do not currently support dagstermill.yield_result"
      )

this is maybe a bit hacky but you could manually write your result to storage, then you could make no-op asset with the name set so that it corresponds with the location where you store the result in the notebook. This asset would take the notebook asset as a

non_argument_deps

. then i think you could use the asset corresponding to the result in other assets and it would fetch the correct value from storage. big caveat - i’ve never tried this and i dont know if you’ll run into typing issues (returning None from the asset but expecting it to be a typed value in a downstream asset may be an issue) or other major issues. If you want to give it a try i recommend writing up a super simple notebook + asset graph to test it out

Thanks @jamie I was thinking the same thing. Thanks for confirming that solution.