I am very new in Python and I read at this moment just the tutorial.
I am confused about the reduceat() function.
I saw the example:
np.add.reduceat([0,1,2,3,4,5,6,7],[0,4,1,5,2,6,3,7])[::2]
and result is:
array([ 6, 10, 14, 18])
How does it come out? Can some one explain for me?
It is sort of like a rolling apply, see:
In [59]:
np.add.reduceat([0,1,2,3,4,5,6,7],[0,4])
Out[59]:
array([ 6, 22])
In [65]:
np.add.reduceat([0,1,2,3,4,5,6,7],[4,1])
Out[65]:
array([ 4, 28])
In [66]:
np.add.reduceat([0,1,2,3,4,5,6,7],[1,5])
Out[66]:
array([10, 18])
In [64]:
np.add.reduceat([0,1,2,3,4,5,6,7],[5,2])
Out[64]:
array([ 5, 27])
In [61]:
np.add.reduceat([0,1,2,3,4,5,6,7],[2,6])
Out[61]:
array([14, 13])
In [67]:
np.add.reduceat([0,1,2,3,4,5,6,7],[6,3])
Out[67]:
array([ 6, 25])
In [62]:
np.add.reduceat([0,1,2,3,4,5,6,7],[3,7])
Out[62]:
array([18, 7])
If you want just the 1st value, you can get it done in just one shot:
In [63]:
np.add.reduceat([0,1,2,3,4,5,6,7],[0,4,1,5,2,6,3,7])
Out[63]:
array([ 6, 4, 10, 5, 14, 6, 18, 7])