Details
-
Type:
Story
-
Status: Done
-
Resolution: Done
-
Fix Version/s: None
-
Component/s: None
-
Labels:None
-
Story Points:2
-
Epic Link:
-
Sprint:DRP S17-2
-
Team:Data Release Production
Description
As part of his PhD thesis work on HSC, José Garmilla produced a star-galaxy classifier and associated supporting code. He has now left the field: we should ensure that his code is not abandoned.
On this ticket, please:
- Track down José's code and any associated documentation;
- Make it available in a repository in the lsst-dm organization.
No changes to the code are required.
Attachments
Activity
Yea, sorry!
Jose's code has been tracked down - it consists of 3 repositories:
1. sgs: Python script for matching HSC data to the HST truth catalog, some input data files, Jose's dissertation, and an example of the SQL queries that should be run on the HSC server. It is located at https://github.com/lsst-dm/sgs
2. sgsupervised: Has the bulk of the code including a script that define the separator (dGauss.py), the code that does the training (supervisedEtl.py), and a script that makes useful plots (plotsScript.py) amongst others. the repository is located at https://github.com/lsst-dm/sgsupervised
3. fsbutler: Code that reads in HSC data, essentially an alternative butler implementation. It is located at https://github.com/lsst-dm/sgs-fsbutler
As requested, the code and all associated documentation has been tracked down and added to repositories on lsst-dm.
Hi - could you please take another look at the review?
Looks good, thank you.
Vishal Kasliwal [X]: You've marked this as "in review", but provided no details on the ticket of what's happened or provided links to code on GitHub for review. Please do that, then send it back when there's actually something here that needs reviewing.