Reputation: 1005

compare an array with two other arrays in python

i have a set of 3 arrays which i would compare to eachother. array a contain a set of values and the values of array b and c should be partly the same.

it is so that if let say a[0] == b[0] than is c[0] always a wrong value

for better explanation i try to show what i mean.

import numpy as np

a = np.array([2, 2, 3, 4])
b = np.array([1, 3, 3, 4])
c = np.array([2, 2, 4, 3])

print(a == b)
# return: [False False True  True]
print(a == c)
# return: [True True False False]

as you can see the from both sets i have two True and two False. so if one of both is true the total should be true. when i do the following i get a single True/ False for an array. and the answers are what i want...

print((a == b).all())
# return: False
print((a == c).all())
# return: False

print(a.all() == (b.all() or c.all()))
# return: True

When i make the array b and c so that i have one vallue that in both cases is wrong i should end with a False

import numpy as np

a = np.array([2, 2, 3, 4])
b = np.array([1, 3, 3, 4])
c = np.array([5, 2, 4, 3])


print(a == b)
# return: [False False True  True]
print(a == c)
# return: [False True False False]

print((a == b).all())
# return: False
print((a == c).all())
# return: False

So far so Good

print(a.all() == (b.all() or c.all()))
# return: True

this part is not good this should be a False!! How do i get an or function like this so that i end with an True when for each value in a an same value exist in b or c??

EDIT: explanation about a[0] == b[0] than is c[0]: i have an python function where phase information comes in and have to do some actions. before that i want check if i have to do with an array of imaginary values or with an array with phase angles. I want to check this before i do something. The problem is the phase angle because the right side is the inverted phase +/- pi. so for every value i have two options. and yes most of the time it is an exclusive or but in the case +/- pi/2 it is not since both are true and that is also fine...

Upvotes: 2

Answers (4)

Amrit

Reputation: 2175

Actually all() function Return True if all elements of the iterable are true (or if the iterable is empty). And in programming language 0 is considered as false and any other integer is considered as true. So here in your array a,b,c every element is non zero and a.all(), b.all() and c.all() they all are seperately returning true so finally you are getting true.

Upvotes: -1

justhalf

Reputation: 9117

From your example and explanation, I guess what you want is:

For each position, exactly one of b or c has the same value as a

If that's the case, that can be done with the following code:

def is_exclusively_jointly_same(a, b, c):
    return np.logical_xor(a==b, a==c).all() # or use ^ below
    # return ((a==b)^(a==c)).all()

The ^ is exclusive or (XOR) operator, which returns True if and only if exactly one of its argument is True.

So the expression (a==b)^(a==c) means either a==b or a==c, but not both. Then the .all() checks whether this is true for all positions in the array.

Examples:

>>> a=np.array([1,2,3,4,5])
>>> b=np.array([1,2,0,0,5])
>>> c=np.array([0,0,3,4,0])
>>> is_exclusively_jointly_same(a, b, c)
True

>>> a=np.array([1,2,3,4,5])
>>> b=np.array([0,2,0,0,5]) # First value both not 1
>>> c=np.array([0,0,3,4,0])
>>> is_exclusively_jointly_same(a, b, c)
False

>>> a=np.array([1,2,3,4,5])
>>> b=np.array([1,2,0,0,5]) # First value both 1
>>> c=np.array([1,0,3,4,0])
>>> is_exclusively_jointly_same(a, b, c)
False

If what you want is that:

For each position, at least one of b and c should have the same value as a

, then you need to change to OR instead of XOR, as follows:

def is_jointly_same(a, b, c):
    return np.logical_or(a==b, a==c).all() # or use | below
    # return ((a==b) | (a==c)).all()

Examples:

>>> a=np.array([1,2,3,4,5])
>>> b=np.array([1,2,0,0,5])
>>> c=np.array([0,0,3,4,0])
>>> is_jointly_same(a, b, c)
True

>>> a=np.array([1,2,3,4,5])
>>> b=np.array([0,2,0,0,5]) # First value both not 1
>>> c=np.array([0,0,3,4,0])
>>> is_jointly_same(a, b, c)
False

>>> a=np.array([1,2,3,4,5])
>>> b=np.array([1,2,0,0,5]) # First value both 1
>>> c=np.array([1,0,3,4,0])
>>> is_jointly_same(a, b, c)
True

The key here is that .all() should be applied once as final aggregator, when each individual values has already been calculated. So when you see that you are applying .all() multiple times, you should be concerned.

Upvotes: 1