I am new to python and created a small function that does a cluster analysis. The quick rundown is I have to compare two arrays a multitude of times, until it no longer changes. For that I have used a while loop, that loops as long as they are not equal, but I find that I get two different results from != and not ==. MWE:
import numpy as np
a = np.array([1,1,1])
b = np.array([1,2,1])
print((a != b).all())
print(not (a == b))
not (a == b) will raise a ValueError because the truth-value of an array with multiple elements is ambiguous.
The way you invert a boolean array in numpy is with the ~ operator:
>>> a != b
array([False, True, False], dtype=bool)
>>> ~ (a == b)
array([False, True, False], dtype=bool)
>>> (~ (a == b)).all() == (a != b).all()
True
The following two expressions are equal. They return true when there is at least one different element.
print((a != b).any())
print(not (a == b).all())
And the following two also give the same result. They return true when every element in the same position is different in the two arrays.
print((a != b).all())
print(not (a == b).any())
to understand more, look at this code , you want to compare d and c but both are arrays so you should compare correctly not by just not operator!!
a = np.array([1,1,1])
b = np.array([1,2,1])
c = (a!=b)
type(c)
Out[6]: numpy.ndarray
type(a)
Out[7]: numpy.ndarray
c
Out[8]: array([False, True, False], dtype=bool)
c.all()
Out[9]: False
d = (a==b)
type(d)
Out[11]: numpy.ndarray
d
Out[12]: array([ True, False, True], dtype=bool)
not d
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-13-84cd3b926fb6> in <module>()
----> 1 not d
ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()
not(d.any())
Out[15]: False
x = d.any()
type(x)
Out[19]: numpy.bool_
not (a == b).all()?