slots and tuple

Learn memory savings with __slots__ and tuple.

What is slots?

__slots__ prevents instances of a class from creating a __dict__. Instead only the specified fields get space; this saves memory.

Normally each instance carries a __dict__ dict; this allows dynamic attribute addition but costs memory. Defining __slots__ = ['x', 'y'] makes instances carry only these fields; no __dict__. When thousands of instances are created the saving matters. But dynamic attributes cannot be added; this is a trade-off.

Measuring memory savings

You check whether an instance has __dict__ with hasattr(obj, '__dict__'). In a class with __slots__ this returns False.

In a normal class it returns True. This is the simplest way to confirm slots is working. To measure the saving you use sys.getsizeof; but __dict__ itself is a separate object and getsizeof does not count it. To see the full saving you must add the __dict__ size too.

Exercise Compare dict with slots

Write a WithSlots class with __slots__ = ['x', 'y']. Write a WithoutSlots class (no slots). Create an instance of each; set x=1, y=2 on both. Print hasattr(a, '__dict__') and hasattr(b, '__dict__') on separate lines.

Show example solution

class WithSlots:
    __slots__ = ["x", "y"]

class WithoutSlots:
    pass

a = WithSlots()
a.x = 1
a.y = 2
b = WithoutSlots()
b.x = 1
b.y = 2
print(hasattr(a, "__dict__"))
print(hasattr(b, "__dict__"))

Test yourself

Match slots concepts

__slots__

Field list
__dict__

Dynamic dict
memory

Saves memory
trade-off

Less flexibility
What does __slots__ prevent?
```
class Point:
    __slots__ = ['x', 'y']
```
- Creating __dict__
- Inheritance
- Method definitions
- Calling __init__
Show answer

Creating __dict__ — __slots__ prevents instances from carrying __dict__; it saves memory.

Complete the __slots__ definition

class Point:
    ____ = ['x', 'y']

Show answer

class Point:
    __slots__ = ['x', 'y']

Next lesson: numpy vectorization

We Are Enhancing Your Experience

Cookie Preferences

Strictly Necessary Cookies

Analytical Cookies