Missing Dataset Type Error

Aysu · January 8, 2025, 2:02am

Hi,

I am following the Tutorials on the LSST Science Pipeline. I followed the steps exactly for each tutorial, except step 3 which was about visualisation. In Tutorial part 5, when I run

pipetask run --register-dataset-types \
-b $RC2_SUBSET_DIR/SMALL_HSC/butler.yaml \
-i u/$USER/warps \
-o u/$USER/coadds \
-p $DRP_PIPE_DIR/pipelines/HSC/DRP-RC2_subset.yaml#assembleCoadd \
-d "skymap = 'hsc_rings_v1' AND tract = 9813 AND patch in (38, 39, 40, 41)" \
-j  7

I get the following error,

 File "/home/lunaeaqua/Research/lsstsw/stack/lsst-scipipe-9.0.0/Linux64/daf_butler/g1ba37b057b+071c5ee9f7/python/lsst/daf/butler/registry/sql_registry.py", line 1917, in _standardize_query_dataset_args
    raise MissingDatasetTypeError(
lsst.daf.butler._exceptions.MissingDatasetTypeError: "Dataset type(s) ['deepCoaddVisits'] are not registered."

How can I solve this?

Aysu · January 9, 2025, 6:07am

I solved it. Apparently one needs to run with #selectDeepCoaddVisits after #makeWrap, before running with #assembleCoadd which is skipped in the documentation.

VenIl · January 16, 2025, 1:58am

Hi,

Could you please elaborate.

Thank you

ryanlau · January 16, 2025, 5:46pm

Hi Ilya, would you mind describing what more information you want about this missing dataset type error?

Perhaps @Aysu might also be able to provide more context to their fix.

Thanks both!

VenIl · January 16, 2025, 5:59pm

Hi, I’m confused by " run with #selectDeepCoaddVisits after #makeWrap, before running with #assembleCoadd which is skipped in the documentation. ". is this about running pipestack once or is this about different steps?

Thank you.

Aysu · January 16, 2025, 6:56pm

Hi, I’m sorry, I’ve been busy in AAS this week. I am traveling back today, I will try to answer you as soon as possible.

Aysu · January 18, 2025, 4:47pm

Between these two pipesask run in the screenshot, I had to run the first one with #selectDeepCoaddVisits before running the second one.

ryanlau · January 18, 2025, 6:31pm

Thanks, Aysu! @VenIl , I hope that addresses your inquiry.

VenIl · January 18, 2025, 7:58pm

Thank you for your response.