Check for NaNs and raise warning

[ayoubft]

As discussed with @observingClouds and @milankl, and raised in this BitInformation.jl issue: the elements in an array should be checked whether they are finite and raise a warning if they are not. As having many NaNs in the dataset may yield unexpected results.

A check could be using is_null() from xarray like:

ds.is_null().sum()

Or perhaps by iterating through a dataset and issuing a warning upon encountering the first NaN value.

[observingClouds]

ds.is_null().any() is probably the way to go. The function is embarrassingly parallel. This shouldn't have much influence on the performance. I probably would call this when we need to load the data anyway.

[ambiguousphoton]

If this issue is still open please assign it to me, I would like to work on it.

[observingClouds]

Hi @ambiguousphoton,
Thank you for your interest in this project. I just assigned you to this issue. I'm looking forward to review your contribution and am happy to help.
Cheers!

[Shaheer-Ahmd]

@observingClouds I would love to implement this.
Can you please guide me about where is this check needed in the codebase?

[observingClouds]

@Shaheer-Ahmd sry for the late response. If you are still interested, the implementation could be added before

xbitinfo/xbitinfo/xbitinfo.py

Line 180 in 7341975

if dim is None and axis is None:

. Please note though that we also want to have tests implemented for this feature.

[observingClouds]

unassigned @ambiguousphoton due to inactivity

[observingClouds]

This issue will be fixed by #260