group similar entries python

Question

I have a numpy array:

array([  2.86656000e+09,   2.86688000e+09,   2.86708000e+09,
     2.86860000e+09,   2.86884000e+09,   2.86908000e+09,
     2.86920000e+09,   2.87024000e+09,   2.87040000e+09,
     2.87056000e+09,   2.87076000e+09,   2.87108000e+09,
     2.87120000e+09,   2.87152000e+09,   2.87260000e+09,
     2.87272000e+09,   2.87280000e+09,   2.87448000e+09,
     2.87464000e+09,   2.87476000e+09,   2.87484000e+09])

what is the best way to group similar values (difference not bigger as 1000000 for example)? Thanks for answers!!

How result should look like? Do you need one group or multiple? If multiple can they intersect? — Igonato, Feb 10 '13 at 20:58
indeed, didn't see that. Thanks exactly what I was looking for... — wa4557, Feb 10 '13 at 21:09

root · Accepted Answer · 2013-02-10T22:41:05.583

As an alternative to the solution linked in the comments, something like this should work:

In [1]: import numpy as np

In [2]: arr = np.array([  2.86656000e+09,   2.86688000e+09,   2.86708000e+09,
   ...:      2.86860000e+09,   2.86884000e+09,   2.86908000e+09,
   ...:      2.86920000e+09,   2.87024000e+09,   2.87040000e+09,
   ...:      2.87056000e+09,   2.87076000e+09,   2.87108000e+09,
   ...:      2.87120000e+09,   2.87152000e+09,   2.87260000e+09,
   ...:      2.87272000e+09,   2.87280000e+09,   2.87448000e+09,
   ...:      2.87464000e+09,   2.87476000e+09,   2.87484000e+09])

In [3]: np.split(arr, np.where(np.diff(arr) > 1000000)[0] + 1)
Out[3]: 
[array([  2.86656000e+09,   2.86688000e+09,   2.86708000e+09]),
 array([  2.86860000e+09,   2.86884000e+09,   2.86908000e+09,
         2.86920000e+09]),
 array([  2.87024000e+09,   2.87040000e+09,   2.87056000e+09,
         2.87076000e+09,   2.87108000e+09,   2.87120000e+09,
         2.87152000e+09]),
 array([  2.87260000e+09,   2.87272000e+09,   2.87280000e+09]),
 array([  2.87448000e+09,   2.87464000e+09,   2.87476000e+09,
         2.87484000e+09])]

group similar entries python

1 Answers1