In_DESI code. about desimodel

[dkirkby]

Is this a duplicate of #25?

[martinjameswhite]

David, Closely related, but I don't think a full duplicate. Stephen requested some code which would do the footprint directly based on the list of tile centers, just applying the "am I in a disc of 1.6 degrees from this center" test. I already have C++ code to do this for the mocks (in MockingDESI) but the issue was more "packaging" and Python compatability. Stephen wanted something easily callable from Python as in: yesno = isindesi(ra, dec, tiles=None) where ra,dec are arrays, and yesno is either a boolean array with same length as ra,dec or alternatively an array of indices where ra,dec are in the desi footprint. If tiles==None, load from desimodel.io.load_tiles(); otherwise take a table of tile centers to use (useful for alternate tiling studies, footprint subset tests, etc). I volunteered Yu to help me to do this ;). Martin.

…

On Tue, Jan 3, 2017 at 4:45 PM, David Kirkby ***@***.***> wrote: Is this a duplicate of #25 <#25>? — You are receiving this because you were assigned. Reply to this email directly, view it on GitHub <#35 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AE-_B9JWbVU2zRUxSPb5yIPmYWKFMy98ks5rOuvCgaJpZM4LaF--> .

[rncahn]

Martin, How precisely defined is "in the footprint"? Does that mean a fiber can actually reach it? Bob

…

On Tue, Jan 3, 2017 at 2:23 PM, Martin White ***@***.***> wrote: Write a piece of code to return whether or not a set of RA/DEC lies within the DESI footprint using the "official" DESI tiles. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#35>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AIeo3bcC3bL8ALsERQ0nGbc4XRsI55dvks5rOspHgaJpZM4LaF--> .

[martinjameswhite]

Bob,

A point is defined to be within the footprint if it is within 1.6 degrees of a tile center (as contained in the "official" FITS file). It does not mean that a fiber can actually reach it, or that the imaging will have sufficient depth, or it is far enough from the nearest bright star or whatever.

Martin.

[martinjameswhite]

When we wrote the mock making code for BOSS (and when we implemented mksample for that matter) we defined in_survey to return not a boolean but rather an integer. This was modeled on the old Unix convention that returning 0 meant "success" and higher numbers indicated what went wrong -- for example not in the footprint, too close to a bright object, "scratches" in the imaging, etc. The code ran through a series of veto masks to populate this return value. I'm not sure whether there is interest in such a thing here, but I thought I'd raise the "issue".

[rainwoodman]

Martin, I imagine another function that combines the output of this and a variety of other functions (too close to a bright object, scratches in the imaging etc) into a bitmask. Stephen already has some sort of mask like this.

[dkirkby]

I sent Stephen (and others) the notebook mentioned in #25 with the relevant code just before the OSU meeting, but we were all preoccupied with final checks on the end-to-end data challenge simulations. I will copy the relevant code to #37 today.

[sbailey]

Fixed via PR #37.