Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-6367

Clarify role of DM in producing/serving photometric redshifts

    Details

    • Team:
      DM Science

      Description

      The DPDD mentions storage of photometric redshifts but gives no indication of where they will come from or how they will be produced. I realize we have perhaps intentionally punted on this issue in the past, and we expect outsiders to do the work, but we now need to have the procedures for accepting, validating (or not), and loading photometric redshifts better defined in order to adequately plan.

        Attachments

          Issue Links

            Activity

            Hide
            mgraham Melissa Graham added a comment -

            After the DM-SST discussion on Mon Dec 2, several changes were made to DMTN-049 (most recent PDF here):

            (1) The timeline dates were fixed so that the community feedback items end in Q2 2020 (minimum attributes) and Q4 2020 (evaluation of options), and the implementation process ends in Q4 2022 (i.e., the end of construction and start of operations).

            (2) All "estimated levels of effort" were removed, as they were only confusing.

            (3) As requested, more detailed description of the core steps of the implementation process are now laid out in Section 5. As discussed, since this is an SST document, it does not contain technical details, and focuses instead on the roles of the commissioning data and community contributions to implementation.

            (4) The long and detailed section which contained the results from several photo-z algorithms that I ran on LSST-like data has been removed – it is unnecessary now that Schmidt et al. will soon be published (and other examples of photo-z estimator comparison papers are given in Section 4.1 anyway).

            The next things to happen should be:

            (1) DM-SST should review the new text of DMTN-049 in full. At minimum Leanne Guy and Jim Bosch I think should read it, and if Keith Bechtol could review the commissioning-related statements, that would be great

            (2) If DM-SST agrees it is necessary, MLG can write an LCR to update DMS-REQ-0046 as described in the above comment from 06/Nov/19. But it's a very minor change, perhaps not worth it.

            (3) We close this ticket and open two new ones to solicit feedback regarding the minimum attributes (due June 30 2020) and the evaluation of potential algorithms (due Dec 31 2020). I start by sending DMTN-049 to the participants of the PCW's session on photo-z, and then to the DESC PZ working group, and then expand more broadly to solicit input from the science collaborations.

            Show
            mgraham Melissa Graham added a comment - After the DM-SST discussion on Mon Dec 2, several changes were made to DMTN-049 ( most recent PDF here ): (1) The timeline dates were fixed so that the community feedback items end in Q2 2020 (minimum attributes) and Q4 2020 (evaluation of options), and the implementation process ends in Q4 2022 (i.e., the end of construction and start of operations). (2) All "estimated levels of effort" were removed, as they were only confusing. (3) As requested, more detailed description of the core steps of the implementation process are now laid out in Section 5. As discussed, since this is an SST document, it does not contain technical details, and focuses instead on the roles of the commissioning data and community contributions to implementation. (4) The long and detailed section which contained the results from several photo-z algorithms that I ran on LSST-like data has been removed – it is unnecessary now that Schmidt et al. will soon be published (and other examples of photo-z estimator comparison papers are given in Section 4.1 anyway). The next things to happen should be: (1) DM-SST should review the new text of DMTN-049 in full. At minimum  Leanne Guy and Jim Bosch I think should read it, and if Keith Bechtol could review the commissioning-related statements, that would be great (2) If DM-SST agrees it is necessary, MLG can write an LCR to update DMS-REQ-0046 as described in the above comment from 06/Nov/19. But it's a very minor change, perhaps not worth it. (3) We close this ticket and open two new ones to solicit feedback regarding the minimum attributes (due June 30 2020) and the evaluation of potential algorithms (due Dec 31 2020). I start by sending DMTN-049 to the participants of the PCW's session on photo-z, and then to the DESC PZ working group, and then expand more broadly to solicit input from the science collaborations.
            Hide
            mgraham Melissa Graham added a comment -

            Extended due date from end of Jan 2020 to near the end of April 2020 (four weeks after the end of the algorithms workshop).

            Show
            mgraham Melissa Graham added a comment - Extended due date from end of Jan 2020 to near the end of April 2020 (four weeks after the end of the algorithms workshop).
            Hide
            mgraham Melissa Graham added a comment -

            Consider adding citations to Nishizawa et al. (2020), "Photometric Redshifts for the Hyper Suprime-Cam Subaru Strategic Program Data Release 2" https://arxiv.org/abs/2003.01511

            Show
            mgraham Melissa Graham added a comment - Consider adding citations to Nishizawa et al. (2020), "Photometric Redshifts for the Hyper Suprime-Cam Subaru Strategic Program Data Release 2"  https://arxiv.org/abs/2003.01511
            Hide
            mgraham Melissa Graham added a comment -

            Mon Aug 17 – put this ticket back "In Progress" since review never completed and DMTN-049's proposed timelines are now out of date. DMTN-049 will be revised and re-presented to the DM-SST on Mon Aug 24.

            Show
            mgraham Melissa Graham added a comment - Mon Aug 17 – put this ticket back "In Progress" since review never completed and DMTN-049's proposed timelines are now out of date. DMTN-049 will be revised and re-presented to the DM-SST on Mon Aug 24.
            Hide
            lguy Leanne Guy added a comment -

            As discussed at the DM SST meeting (2020-08-17) and linked to this ticket. We should add a section to this document on how we expect to work with any in-kind contribution to photo-z and open questions such as testing and long-term support.

            Show
            lguy Leanne Guy added a comment - As discussed at the DM SST meeting (2020-08-17) and linked to this ticket. We should add a section to this document on how we expect to work with any in-kind contribution to photo-z and open questions such as testing and long-term support.

              People

              • Assignee:
                mgraham Melissa Graham
                Reporter:
                jbosch Jim Bosch
                Reviewers:
                Jim Bosch, Keith Bechtol, Leanne Guy
                Watchers:
                Andrew Connolly, Eli Rykoff, Gregory Dubois-Felsmann, Jacek Becla, Jim Bosch, John Swinbank, Jonathan Sick, Keith Bechtol, Leanne Guy, Melissa Graham, Tim Jenness, Wil O'Mullane, Zeljko Ivezic
              • Votes:
                0 Vote for this issue
                Watchers:
                13 Start watching this issue

                Dates

                • Due:
                  Created:
                  Updated:
                  End date:

                  Summary Panel