26

I'm trying to find an elegant way to find the max value in a two-dimensional array. for example for this array:

[0, 0, 1, 0, 0, 1] [0, 1, 0, 2, 0, 0][0, 0, 2, 0, 0, 1][0, 1, 0, 3, 0, 0][0, 0, 0, 0, 4, 0]

I would like to extract the value '4'. I thought of doing a max within max but I'm struggling in executing it.

snakecharmerb
  • 36,887
  • 10
  • 71
  • 115
Shuki
  • 263
  • 1
  • 3
  • 4

6 Answers6

33

Another way to solve this problem is by using function numpy.amax()

>>> import numpy as np
>>> arr = [0, 0, 1, 0, 0, 1] , [0, 1, 0, 2, 0, 0] , [0, 0, 2, 0, 0, 1] , [0, 1, 0, 3, 0, 0] , [0, 0, 0, 0, 4, 0]
>>> np.amax(arr)
Khan Saad Bin Hasan
  • 460
  • 6
  • 7
29

Max of max numbers (map(max, numbers) yields 1, 2, 2, 3, 4):

>>> numbers = [0, 0, 1, 0, 0, 1], [0, 1, 0, 2, 0, 0], [0, 0, 2, 0, 0, 1], [0, 1, 0, 3, 0, 0], [0, 0, 0, 0, 4, 0]

>>> map(max, numbers)
<map object at 0x0000018E8FA237F0>
>>> list(map(max, numbers))  # max numbers from each sublist
[1, 2, 2, 3, 4]

>>> max(map(max, numbers))  # max of those max-numbers
4
falsetru
  • 336,967
  • 57
  • 673
  • 597
10

Not quite as short as falsetru's answer but this is probably what you had in mind:

>>> numbers = [0, 0, 1, 0, 0, 1], [0, 1, 0, 2, 0, 0], [0, 0, 2, 0, 0, 1], [0, 1, 0, 3, 0, 0], [0, 0, 0, 0, 4, 0]
>>> max(max(x) for x in numbers)
4
schmy
  • 101
  • 1
  • 4
3

How about this? 

import numpy as np
numbers = np.array([[0, 0, 1, 0, 0, 1], [0, 1, 0, 2, 0, 0], [0, 0, 2, 0, 0, 1], [0, 1, 0, 3, 0, 0], [0, 0, 0, 0, 4, 0]])

print(numbers.max())

4
cebe
  • 3,230
  • 1
  • 24
  • 37
Gurami
  • 31
  • 1
2
>>> numbers = [0, 0, 1, 0, 0, 1], [0, 1, 0, 2, 0, 0], [0, 0, 2, 0, 0, 1], [0, 1, 0, 3, 0, 0], [0, 0, 0, 0, 4, 0]

You may add key parameter to max as below to find Max value in a 2-D Array/List

>>> max(max(numbers, key=max))
4
afghani
  • 69
  • 1
  • 5
1

One very easy solution to get both the index of your maximum and your maximum could be :

numbers = np.array([[0,0,1,0,0,1],[0,1,0,2,0,0],[0,0,2,0,0,1],[0,1,0,3,0,0],[0,0,0,0,4,0]])
ind = np.argwhere(numbers == numbers.max()) # In this case you can also get the index of your max
numbers[ind[0,0],ind[0,1]]
Markh
  • 63
  • 5