Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-16807

Compare HiTS output to Martinez 2018

    Details

    • Story Points:
      12
    • Epic Link:
    • Team:
      Alert Production

      Description

      http://adsabs.harvard.edu/abs/2018AJ....156..186M produced classified lightcurves for the HiTS dataset. This ticket is to produce an initial comparison of that catalog to the results of our ap_pipe processing.

        Attachments

          Activity

          Hide
          ebellm Eric Bellm added a comment -

          Upon examination, the catalog produced by the paper above appears to be highly incomplete--their machine learning procedure appears to only select very high-confidence variables. It is probably worth proceeding with this ticket to get some example lightcurves to refer back to, but we should do so with the understanding that the catalog does not provide a "truth table" in any sense.

          Show
          ebellm Eric Bellm added a comment - Upon examination, the catalog produced by the paper above appears to be highly incomplete--their machine learning procedure appears to only select very high-confidence variables. It is probably worth proceeding with this ticket to get some example lightcurves to refer back to, but we should do so with the understanding that the catalog does not provide a "truth table" in any sense.
          Hide
          ebellm Eric Bellm added a comment -

          On communication with the author of the ticket, we obtained the features table that has the fit periods, etc. However, surprisingly the internal IDs of the features do not match those of the objects, which limits their utility.

          Show
          ebellm Eric Bellm added a comment - On communication with the author of the ticket, we obtained the features table that has the fit periods, etc. However, surprisingly the internal IDs of the features do not match those of the objects, which limits their utility.
          Hide
          ebellm Eric Bellm added a comment -

          Resolved the internal ID problem with the author.

          Show
          ebellm Eric Bellm added a comment - Resolved the internal ID problem with the author.
          Hide
          ebellm Eric Bellm added a comment -

          Hi Meredith, can you take a look at this notebook, which gives us a sample of 44 high-confidence variable stars in our HiTS fields?

          Show
          ebellm Eric Bellm added a comment - Hi Meredith, can you take a look at this notebook, which gives us a sample of 44 high-confidence variable stars in our HiTS fields?
          Hide
          mrawls Meredith Rawls added a comment -

          This is great! Thanks for corresponding with the HiTS DR folks to get the correct features table.

          Please document where the two FITS catalogs you made at the end are living, and please also add a few more comments/captions/axis labels in the intermediate figures and data explorations throughout the notebook. That will make our future-selves a little less unhappy, I suspect. I'll go ahead and mark this as Reviewed, but I would like to see those small changes (and a PR so ap_pipe-notebooks master is up-to-date) before it is officially Done.

          At the end of this, there are some 150ish variable sources, 44 of which are "high confidence variable stars," yes? I'm looking forward to seeing if we can verify these and maybe find some more with LSST software!

          Show
          mrawls Meredith Rawls added a comment - This is great! Thanks for corresponding with the HiTS DR folks to get the correct features table. Please document where the two FITS catalogs you made at the end are living, and please also add a few more comments/captions/axis labels in the intermediate figures and data explorations throughout the notebook. That will make our future-selves a little less unhappy, I suspect. I'll go ahead and mark this as Reviewed, but I would like to see those small changes (and a PR so ap_pipe-notebooks master is up-to-date) before it is officially Done. At the end of this, there are some 150ish variable sources, 44 of which are "high confidence variable stars," yes? I'm looking forward to seeing if we can verify these and maybe find some more with LSST software!
          Hide
          ebellm Eric Bellm added a comment -

          Meredith Rawls, the two labeled variable sets are now included in the data/ directory of ap_pipe-notebooks.

          Show
          ebellm Eric Bellm added a comment - Meredith Rawls , the two labeled variable sets are now included in the data/ directory of ap_pipe-notebooks .

            People

            • Assignee:
              ebellm Eric Bellm
              Reporter:
              ebellm Eric Bellm
              Reviewers:
              Meredith Rawls
              Watchers:
              Eric Bellm, Meredith Rawls
            • Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Summary Panel