Adding rubin_proc

[bungerbuilder]

Rubin_proc, given a visitId and detector number, will in the RSP DP1 environment:

• retrieves a visit_image using a visit ID and detector
• converts the Rubin PSF into a Crowdsource-compatible PSF
• prepares the image, inverse-uncertainty, and mask arrays
• passes processing options to crowdsource_base.fit_im
• returns both the Crowdsource results and Rubin exposure

[schlafly]

Some minor comments in line, but this looks pretty good, thanks!

[schlafly]

We'll want to avoid relying on WISE here, since that will do some weird things, like assuming that the diffraction spikes get spread out near the ecliptic poles.

We probably want to eventually just overhaul the PSF determination, but it's also worth thinking about what the right behavior is wrt negatives in the Rubin PSF; I doubt we 'really' want to clip and renormalize, but it's fine for now.

[schlafly]

Please add a CLI along the lines of what is in simple_proc, so that you can either use this from python or from the command line. That's the if name == 'main' thing at the end.

[schlafly]

Just out of curiosity, have you tried the CLI? It looks right to me but it would be good to try it once.

[schlafly]

You mentioned needing an update to psf.py here, but I don't see it. You'll probably need to git add that file.

[bungerbuilder]

Rubin_proc, given a visitId and detector number, will in the RSP DP1 environment:

• retrieves a visit_image using a visit ID and detector
• converts the Rubin PSF into a Crowdsource-compatible PSF
• prepares the image, inverse-uncertainty, and mask arrays
• passes processing options to crowdsource_base.fit_im
• returns both the Crowdsource results and Rubin exposure

[schlafly]

Some minor comments in line, but this looks pretty good, thanks!

[schlafly]

We'll want to avoid relying on WISE here, since that will do some weird things, like assuming that the diffraction spikes get spread out near the ecliptic poles.

We probably want to eventually just overhaul the PSF determination, but it's also worth thinking about what the right behavior is wrt negatives in the Rubin PSF; I doubt we 'really' want to clip and renormalize, but it's fine for now.

[schlafly]

Please add a CLI along the lines of what is in simple_proc, so that you can either use this from python or from the command line. That's the if name == 'main' thing at the end.

[schlafly]

Just out of curiosity, have you tried the CLI? It looks right to me but it would be good to try it once.

[schlafly]

You mentioned needing an update to psf.py here, but I don't see it. You'll probably need to git add that file.

[bungerbuilder]

Rubin_proc, given a visitId and detector number, will in the RSP DP1 environment:

• retrieves a visit_image using a visit ID and detector
• converts the Rubin PSF into a Crowdsource-compatible PSF
• prepares the image, inverse-uncertainty, and mask arrays
• passes processing options to crowdsource_base.fit_im
• returns both the Crowdsource results and Rubin exposure

[schlafly]

Some minor comments in line, but this looks pretty good, thanks!

[schlafly]

We'll want to avoid relying on WISE here, since that will do some weird things, like assuming that the diffraction spikes get spread out near the ecliptic poles.

We probably want to eventually just overhaul the PSF determination, but it's also worth thinking about what the right behavior is wrt negatives in the Rubin PSF; I doubt we 'really' want to clip and renormalize, but it's fine for now.

[schlafly]

Please add a CLI along the lines of what is in simple_proc, so that you can either use this from python or from the command line. That's the if name == 'main' thing at the end.

[schlafly]

Just out of curiosity, have you tried the CLI? It looks right to me but it would be good to try it once.

[schlafly]

You mentioned needing an update to psf.py here, but I don't see it. You'll probably need to git add that file.

[bungerbuilder]

Rubin_proc, given a visitId and detector number, will in the RSP DP1 environment:

• retrieves a visit_image using a visit ID and detector
• converts the Rubin PSF into a Crowdsource-compatible PSF
• prepares the image, inverse-uncertainty, and mask arrays
• passes processing options to crowdsource_base.fit_im
• returns both the Crowdsource results and Rubin exposure

[schlafly]

Some minor comments in line, but this looks pretty good, thanks!

[schlafly]

We'll want to avoid relying on WISE here, since that will do some weird things, like assuming that the diffraction spikes get spread out near the ecliptic poles.

We probably want to eventually just overhaul the PSF determination, but it's also worth thinking about what the right behavior is wrt negatives in the Rubin PSF; I doubt we 'really' want to clip and renormalize, but it's fine for now.

[schlafly]

Please add a CLI along the lines of what is in simple_proc, so that you can either use this from python or from the command line. That's the if name == 'main' thing at the end.

[schlafly]

Just out of curiosity, have you tried the CLI? It looks right to me but it would be good to try it once.

[schlafly]

You mentioned needing an update to psf.py here, but I don't see it. You'll probably need to git add that file.

[bungerbuilder]

Rubin_proc, given a visitId and detector number, will in the RSP DP1 environment:

• retrieves a visit_image using a visit ID and detector
• converts the Rubin PSF into a Crowdsource-compatible PSF
• prepares the image, inverse-uncertainty, and mask arrays
• passes processing options to crowdsource_base.fit_im
• returns both the Crowdsource results and Rubin exposure

[schlafly]

Some minor comments in line, but this looks pretty good, thanks!

[schlafly]

We'll want to avoid relying on WISE here, since that will do some weird things, like assuming that the diffraction spikes get spread out near the ecliptic poles.

We probably want to eventually just overhaul the PSF determination, but it's also worth thinking about what the right behavior is wrt negatives in the Rubin PSF; I doubt we 'really' want to clip and renormalize, but it's fine for now.

[schlafly]

Please add a CLI along the lines of what is in simple_proc, so that you can either use this from python or from the command line. That's the if name == 'main' thing at the end.

[schlafly]

Just out of curiosity, have you tried the CLI? It looks right to me but it would be good to try it once.

[schlafly]

You mentioned needing an update to psf.py here, but I don't see it. You'll probably need to git add that file.

[bungerbuilder]

Rubin_proc, given a visitId and detector number, will in the RSP DP1 environment:

• retrieves a visit_image using a visit ID and detector
• converts the Rubin PSF into a Crowdsource-compatible PSF
• prepares the image, inverse-uncertainty, and mask arrays
• passes processing options to crowdsource_base.fit_im
• returns both the Crowdsource results and Rubin exposure

[schlafly]

Some minor comments in line, but this looks pretty good, thanks!

[schlafly]

We'll want to avoid relying on WISE here, since that will do some weird things, like assuming that the diffraction spikes get spread out near the ecliptic poles.

We probably want to eventually just overhaul the PSF determination, but it's also worth thinking about what the right behavior is wrt negatives in the Rubin PSF; I doubt we 'really' want to clip and renormalize, but it's fine for now.

[schlafly]

Please add a CLI along the lines of what is in simple_proc, so that you can either use this from python or from the command line. That's the if name == 'main' thing at the end.

[schlafly]

Just out of curiosity, have you tried the CLI? It looks right to me but it would be good to try it once.

[schlafly]

You mentioned needing an update to psf.py here, but I don't see it. You'll probably need to git add that file.

[bungerbuilder]

Rubin_proc, given a visitId and detector number, will in the RSP DP1 environment:

• retrieves a visit_image using a visit ID and detector
• converts the Rubin PSF into a Crowdsource-compatible PSF
• prepares the image, inverse-uncertainty, and mask arrays
• passes processing options to crowdsource_base.fit_im
• returns both the Crowdsource results and Rubin exposure

[schlafly]

Some minor comments in line, but this looks pretty good, thanks!

[schlafly]

We'll want to avoid relying on WISE here, since that will do some weird things, like assuming that the diffraction spikes get spread out near the ecliptic poles.

We probably want to eventually just overhaul the PSF determination, but it's also worth thinking about what the right behavior is wrt negatives in the Rubin PSF; I doubt we 'really' want to clip and renormalize, but it's fine for now.

[schlafly]

Please add a CLI along the lines of what is in simple_proc, so that you can either use this from python or from the command line. That's the if name == 'main' thing at the end.

[schlafly]

Just out of curiosity, have you tried the CLI? It looks right to me but it would be good to try it once.

[schlafly]

You mentioned needing an update to psf.py here, but I don't see it. You'll probably need to git add that file.

[bungerbuilder]

Rubin_proc, given a visitId and detector number, will in the RSP DP1 environment:

• retrieves a visit_image using a visit ID and detector
• converts the Rubin PSF into a Crowdsource-compatible PSF
• prepares the image, inverse-uncertainty, and mask arrays
• passes processing options to crowdsource_base.fit_im
• returns both the Crowdsource results and Rubin exposure

[schlafly]

Some minor comments in line, but this looks pretty good, thanks!

[schlafly]

We'll want to avoid relying on WISE here, since that will do some weird things, like assuming that the diffraction spikes get spread out near the ecliptic poles.

We probably want to eventually just overhaul the PSF determination, but it's also worth thinking about what the right behavior is wrt negatives in the Rubin PSF; I doubt we 'really' want to clip and renormalize, but it's fine for now.

[schlafly]

Please add a CLI along the lines of what is in simple_proc, so that you can either use this from python or from the command line. That's the if name == 'main' thing at the end.

[schlafly]

Just out of curiosity, have you tried the CLI? It looks right to me but it would be good to try it once.

[schlafly]

You mentioned needing an update to psf.py here, but I don't see it. You'll probably need to git add that file.