Numpy np. Max and np Maximum implements the relu function

1. np.max Calculate the maximum value in an array

（1） Basic usage

a = np.array([
    [1, 2, 3],
    [4, 5, 6],
    [7, 8, 9]
])

print(np.max(a))
print(np.max(a, axis=0))
print(np.max(a, axis=1))

[Out]: 
9
[7 8 9]
[3 6 9]

（2）keepdims： Maintain shape

print(np.max(a, keepdims=True))
print(np.max(a, axis=0, keepdims=True))
print(np.max(a, axis=1, keepdims=True))

[Out]:
[[9]]
[[7 8 9]]
[[3]
 [6]
 [9]]

（3）initial： Specify the minimum

print(np.max(a, initial=8))
print(np.max(a, axis=0, initial=8))
print(np.max(a, axis=1, initial=8))

[Out]:
9
[8 8 9]
[8 8 9]

（4）where： Specify the elements that participate in the comparison , Non participating settings initial value

print(np.max(a, initial=8, where=[1, 1, 0]))
print(np.max(a, axis=0, initial=8, where=[True, True, False]))
print(np.max(a, axis=1, initial=8, where=[True, True, False]))

[Out]:
8
[8 8 8]
[8 8 8]

2. np.maximum Compare the maximum values of two arrays

a = np.array([
    [1, 2, 3],
    [4, 5, 6]
])
b = np.array([
    [7, 8, 9],
    [0, 1, 2]
])

print(np.maximum(a, b))
print(np.maximum(a, 3.5))

[Out]:
[[7 8 9]
 [4 5 6]]
[[3.5 3.5 3.5]
 [4.  5.  6. ]]

def relu(inX):
    return np.maximum(0,inX)