Addition of list and NumPy number

Question

If you add an integer to a list, you get an error raised by the __add__ function of the list (I suppose):

>>> [1,2,3] + 3
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: can only concatenate list (not "int") to list

If you add a list to a NumPy array, I assume that the __add__ function of the NumPy array converts the list to a NumPy array and adds the lists

>>> np.array([3]) + [1,2,3]
array([4, 5, 6])

But what happens in the following?

>>> [1,2,3] + np.array([3])
array([4, 5, 6])

How does the list know how to handle addition with NumPy arrays?

"How does the list know how to handle addition with numpy arrays." - it doesn't, [1,2,3].__add__(np.array([3])) will fail, but Python reverses the arguments if the first attempt fails and np.array([3]).__radd__([1,2,3]) (__add__ is called if __radd__ isn't defined) works. — jonrsharpe
– jonrsharpe, Commented Oct 15, 2015 at 14:32
@jonrsharpe: I don't think a + b will ever call b.__add__. Python doesn't assume that addition is commutative. You may be thinking of the augmented assignment operators, where a += b will first try a.__iadd__, but then fall back to __add__ (and further to b.__radd__) if necessary. — Blckknght
– Blckknght, Commented Oct 16, 2015 at 4:07
@Blckknght you're absolutely right, I guess that's what I was thinking of. — jonrsharpe
– jonrsharpe, Commented Oct 16, 2015 at 7:19

Peter Mortensen · Accepted Answer · 2015-10-15 16:20:00Z

12

list does not know how to handle addition with NumPy arrays. Even in [1,2,3] + np.array([3]), it's NumPy arrays that handle the addition.

As documented in the data model:

For objects x and y, first x.__op__(y) is tried. If this is not implemented or returns NotImplemented, y.__rop__(x) is tried. If this is also not implemented or returns NotImplemented, a TypeError exception is raised. But see the following exception:

Exception to the previous item: if the left operand is an instance of a built-in type or a new-style class, and the right operand is an instance of a proper subclass of that type or class and overrides the base’s __rop__() method, the right operand’s __rop__() method is tried before the left operand’s __op__() method.

When you do

[1,2,3] + np.array([3])

what is internally called is

np.array([3]).__radd__([1,2,3])

edited Oct 15, 2015 at 16:20

Peter Mortensen

31.4k22 gold badges110 silver badges134 bronze badges

answered Oct 15, 2015 at 14:33

Anand S Kumar

91.5k18 gold badges196 silver badges179 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Warren Weckesser · Accepted Answer · 2015-10-15 17:41:02Z

8

It is because of the __radd__ method of np.array, check out this link : http://www.rafekettler.com/magicmethods.html#numeric (paragraph Reflected arithmetic operators).

In facts, when you try [1,2,3].__add__(np.array([3])), it raises an error, so np.array([3]).__radd__([1,2,3]) is called.

edited Oct 15, 2015 at 17:41

Warren Weckesser

116k20 gold badges207 silver badges224 bronze badges

answered Oct 15, 2015 at 14:41

Labo

2,8105 gold badges22 silver badges41 bronze badges

Collectives™ on Stack Overflow

Addition of list and NumPy number

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related