Fred Moolekamp, can you see if this works for you without having to setup/install gauss2d/(meas_extensions_)multiprofit? I tested by including only modelling_research in my PYTHONPATH and it seemed to work.

If you need to compare, I have w44 and w40 notebooks (the latter in an old subdirectory since it won't work as-is anymore).

Sure. Sorry, I forgot about this today but I'll test it out tomorrow.

For reference, the command to run the matcher:

pipetask --long-log run -b /repo/dc2 --input "2.2i/runs/test-med-1/w_2021_44/DM-32391,2.2i/truth_summary" --output "u/dtaranu/DM-32034/w_2021_44_match_0.5asec" -d "instrument = 'LSSTCam-imSim' and tract=3829 and skymap='DC2'" -p /project/dtaranu/dc2/match_catalog_pipe/match_tract_catalog_probabilistic.yaml --register-dataset-types

Something isn't right. When I run the notebook before using this task I get

for the 2nd and 3rd plots in the notebook, but if I use the notebook contained in this ticket I get

when run on the same data.

The task command that I used to merge the catalogs was

pipetask --long-log run -b /repo/dc2 --output-run "u/fred3m/DM-32436/DC2_44/20211210T214832Z" --extend-run -d "instrument = 'LSSTCam-imSim' and tract in (3828,3829) and skymap='DC2'" -p /project/dtaranu/dc2/match_catalog_pipe/match_tract_catalog_probabilistic.yaml --input 2.2i/runs/test-med-1/w_2021_44/DM-32391,2.2i/truth_summary --no-versions

Ok, thanks to assistance from Dan Taranu in slack we were able to get the notebook working.

The changes work to recreate the plots from the previous version of the notebook.