Python中3D数组的最大值

Question

I want to find the max value of a 3d array in python. I tried

image_file1 = open("lena256x256.bmp","rb")
img_i = PIL.Image.open(image_file1)
pix = numpy.array(img_i);
maxval= max(pix)

but i am getting an error

 File "test.py", line 31, in <module>
    maxval= max(pix)
ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

I cannot catch my mistake here, please help me.

Answer 1

You are using the builtin max function that does not understand multidimensional NumPy arrays. You must instead use one of:

• pix.max()
• numpy.max(pix)
• numpy.amax(pix)

These are also faster than the builtin in the case of 1D NumPy arrays.

Answer 2

Max期望使用单个值，错误消息应该非常清楚，您想使用amax代替。

maxval = numpy.amax(pix)

Answer 3

In accordance to what georgesl wrote, you can use flat to get an iterator for the array and then do something like

m = reduce(max, ar.flat)

Edit: removed the lambda , the default max should be OK.

Answer 4

The np.max function works for vectors, not matrices (or along an axis). To have the max element a multi-dimensionnal array, you can use flatten() : maxval= pp.max( pix.flatten() )