numpy 找出最大的数,如何在numpy数组列中找到最大值？

I can find quite a few permutations of this question, but not this (rather simple) one: how do I find the maximum value of a specific column of a numpy array (in the most pythonic way)?

a = array([[10, 2], [3, 4], [5, 6]])

What I want is the max value in the first column and second column (these are x,y coordinates and I eventually need the height and width of each shape), so max x coordinate is 10 and max y coordinate is 6.

I've tried:

xmax = numpy.amax(a,axis=0)

ymax = numpy.amax(a,axis=1)

but these yield

array([10, 6])

array([10, 4, 6])

...not what I expected.

My solution is to use slices:

xmax = numpy.max(a[:,0])

ymax = numpy.max(a[:,1])

Which works but doesn't seem to the best approach.

Suggestions?

解决方案

Just unpack the list:

In [273]: xmax, ymax = a.max(axis=0)

In [274]: print xmax, ymax

#10 6