andykee / prtools Goto Github PK

Describe the issue:

Doctoring says:

nanwarn : bool, optional
        If True, a RuntimeWarning will be raised if NaNs are present
        in the output. Default is False.

The nanwarn argument is never used in the function though.

Reproduce the code example:

n/a

Error message:

No response

Runtime information:

1.0.0
1.24.4
1.10.1
3.10.12 | packaged by conda-forge | (main, Jun 23 2023, 22:41:52) [Clang 15.0.7 ]

Describe the issue:

centroid returns (nan, nan) when provided an input containing NaNs

Reproduce the code example:

import numpy as np
import prtools

c = np.zeros((5,5))
c[2,2] = 1
c[1,1] = np.nan
prtools.centroid(c)

Error message:

No response

Runtime information:

1.0.0
1.22.2
1.8.0
3.8.11 (default, Jul 29 2021, 14:57:32) 
[Clang 12.0.0 ]

Proposed new feature or change:

medfix2 throws a RuntimeWarning if there are large masked areas to repair (larger than kernel size).

The exact message is

/miniconda3/lib/python3.8/site-packages/numpy/lib/nanfunctions.py:1218: RuntimeWarning: All-NaN slice encountered
  r, k = function_base._ureduce(a, func=_nanmedian, axis=axis, out=out, 

Since the function is behaving as expected, it is safe to just suppress this message and carry on with life.

The documentation should be updated to be more explicit about the interpolation behavior in this case.

Proposed new feature or change:

It's often useful to do a quick PSF simulation. If calcpsf is still laying around somewhere, it should be included in this library

Proposed new feature or change:

It would be nice to be able to specify the point about which padding is performed. This is particularly useful when trimming images down to size about a centroid.