fix: load_tiff in README was undefined
fix for #31
Coverage decreased (-0.09%) to 29.905% when pulling 7c81bf62bd05c27eba0b781a7136855257564d2f on culebron:master into e633a4e8d8d168fa2a6b55760f8b1832d3c8b274 on ozak:master.
It may be better to just add **kwargs instead of options=None and options= options, no?
Yes, good idea.
Doesn't forcing a np.float64(ndv) force any dataset to become float, even the ones that are integers and so increasing file size?
@ozak I didn't test this. I found out that integer numpy matrices can't have None in them, and for integer ones (like ASTER DEM) it's only necessary to have a maksed array, without ndv.
It starts being confusing. Let me see how to run tests.
To see the effect of this np.float I would save an integer raster with and without the option to see the difference in size.
I'm not sure I understood what you meant with the last comments... When is None being assigned to a numpy array?
Also, how would you have a masked array without an ndv?