How to invert a masked numpy array?

Question

I've created a mask over a vector shapefile and have then created another one which zooms in to the relevant area. The problem I have with this new mask is that the masked area is the bit that I want, so I need to invert it somehow. (Note that I imported numpy as np)

This is the code I used for the 2nd mask:

catchment_mask = gmask[minrow:minrow+nrow, mincol:mincol+ncol]
catchment_mask = np.where(catchment_mask == 1, 1., np.nan)

I've tried using this code to invert it but it has applied a mask to the entire area:

inverse_catchment_mask = (np.logical_not(catchment_mask))

I'd be really grateful for any suggestions!

Your indentation is off. There is a np.ma, MaskedArray subclass, but I don't think that's what you have. It looks like your mask is a slice of gmask, but then you make a new array using where. Why? That's no longer a view. Why the np.nan? logical_not of that is False. — hpaulj
– hpaulj, Commented Dec 16, 2017 at 19:12

jakevdp · Accepted Answer · 2017-12-16 23:12:53Z

1

If your cachement_mask consisted of boolean (True/False) values, then you solution based on logical_not would work as expected

But your mask consists of 1 or NaN, so you have to invert this more carefully. For example:

inverse_cachement_mask = np.where(np.isnan(cachement_mask), 1, np.nan)

This will create an inverse of your original mask, swapping the 1 and NaN values.

answered Dec 16, 2017 at 23:12

jakevdp

88.5k15 gold badges140 silver badges177 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

JaneOz Over a year ago

Thanks for the explanation. I removed the np.where and changed it for logical_not with the relevant arguments and it has inverted it as expected with boolean values.

Collectives™ on Stack Overflow

How to invert a masked numpy array?

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related