asizeof() gives incorrect values for the memory consumption of named tuples.

[ceridwen]

From the documentation for the collections module:

"Named tuple instances do not have per-instance dictionaries, so they are lightweight and require no more memory than regular tuples."

Python 3.4.0 (default, Apr 11 2014, 13:05:11) 
[GCC 4.8.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import pympler.asizeof
>>> import collections
>>> Point = collections.namedtuple('Point', ['x', 'y'])
>>> p = Point(11, y=22)
>>> p
Point(x=11, y=22)
>>> pympler.asizeof.asizeof(p)
1440
>>> t = 11, 22
>>> t
(11, 22)
>>> pympler.asizeof.asizeof(t)
128

I haven't been able to figure out the reason for the discrepancy.

[Pankrat]

I believe Pympler assumes that the __dict__ property is part of the object and includes it in the given size:

>>> print(asizeof.asized(p, detail=2).format())
Point(x=11, y=22) size=728 flat=32
    __dict__ size=672 flat=576
        [K] x size=32 flat=32
        [K] y size=32 flat=32
        [V] x: 11 size=16 flat=16
        [V] y: 22 size=16 flat=16
    __slots__ size=24 flat=24
    __class__ size=0 flat=0
>>> p.__dict__
OrderedDict([('x', 11), ('y', 22)])

I'd have to check the namedtuple implementation in Python but it really sounds like a bug in Pympler's recursive sizing.