CODEWITHSUNDEEP
pythonarraysnumpyh5py
GeoMonkey
GeoMonkey

Reputation: 1665

subsampling a numpy array?

Im relatively new to numpy but have started using it to read and write from and to h5 files. I have image data on which I have computed some zonal statistics, reading each pixel value in a given zone into a h5 file. However, I have a lot of pixel values (possibly tens of millions) and wanted to subsample this data so that I am able to cut down the data size but keep the general distribution of the data.

I was wondering if there was a simple way of sampling every 200th value of an array?

I would put up what code I have already but my code only goes as far as to read in my existing data - Im completely stuck as to how I might subsample it so have nothing to show so far.

Thanks

Upvotes: 1

Views: 5679

Answers (1)

wim
wim

Reputation: 362488

You can use an array slice:

>>> import numpy as np
>>> a = np.eye(1000)
>>> a[::200, ::200]

array([[ 1.,  0.,  0.,  0.,  0.],
       [ 0.,  1.,  0.,  0.,  0.],
       [ 0.,  0.,  1.,  0.,  0.],
       [ 0.,  0.,  0.,  1.,  0.],
       [ 0.,  0.,  0.,  0.,  1.]])

Upvotes: 8

Related Questions