0

I have a 2d binary array indicating the presence of half-channels at a particular coordinate (0=not present, 1=present). I need to convert this array into a probability distribution to plot on a map of the globe using matplotlib.

I tried dividing each element of the array by the amount of time over which the values are calculated. For example, if the data was taken over a period of one month I divided by 30. I also tried taking the exponent of each value like so:

return np.exp(-x ** 2)

but nothing looks right. Any suggestions? Thanks.

kazooie_tooie
  • 7
  • 7
  • so the 2d binary array contains measurements? if you divide that by the measurement time then you get something like the temporal pdf / cdf for a defined region in your 2d array. Does this help / are you a little bit familiar with stastistics? – Baedsch Oct 11 '18 at 13:44
  • I don't have measurements. I just have 0s and 1s. – kazooie_tooie Oct 16 '18 at 15:31
  • so do you want to perform sth like a [moving average](https://en.wikipedia.org/wiki/Moving_average)? because otherwise everything you can get out of a binary 2d array is sth like a irregular chessboard-like grid, right? – Baedsch Oct 18 '18 at 08:18
  • ah, or do you want to fit multiple [gaussians](https://en.wikipedia.org/wiki/Gaussian_function) onto this "chessboard"? this will return rubbish... – Baedsch Oct 18 '18 at 08:19

0 Answers0