Evaluate grid PSF on a subset?

[keflavich]

How can I run PSF photometry on a cutout of an image using a gridPSF model?

There's probably some clever way to wrap the grid.__call__ to give it an offset, but I haven't come up with it yet.

[keflavich]

This might be as simple as

grid.x_0 = bottom_left_x
grid.y_0 = bottom_left_y

but that didn't work, nor did

def evaluate(x, y, flux, x_0, y_0):
    """
    Evaluate the `GriddedPSFModel` for the input parameters.
    """
    # Get the local PSF at the (x_0,y_0)
    psfmodel = grid._compute_local_model(x_0+slcs[1].start, y_0+slcs[0].start)

    # now evaluate the PSF at the (x_0, y_0) subpixel position on
    # the input (x, y) values
    return psfmodel.evaluate(x, y, flux, x_0, y_0)
grid.evaluate = evaluate

[larrybradley]

Is the cutout small enough that only one ePSF model should be applied, i.e., do you really need an ePSF grid over your cutout?

[keflavich]

For the cutout, I probably only need the one, but I'm really trying to use the cutout as a small-area test region for later extracting photometry generally. I want to work on a cutout to tune the parameters before letting them loose on the full image. So ideally I'd like to keep the same software infrastructure in place.

[larrybradley]

Something like what you did above by shifting the origin should work in principle, but I'd need look for all the places those shifts should be applied.