Optimization of Data Classes

Using Data Classes in Python
Darren Jones 03:17

Transcript
Discussion (1)

00:00 Optimizing data classes. In this section of the course, you’re going to take a look at slots, which can be used to make classes faster and take up less memory.

00:11 Data classes have no explicit syntax for working with slots, but the normal way of creating them works for data classes as well. You probably expect this as, after all, data classes are really just regular classes.

00:59 In essence, slots are defined using the .__slots__ to list the variables on a class. Variables or attributes not present in the .__slots__ list may not be defined. Furthermore, a slots class may not have default values.

01:16 The benefit of adding such restrictions is that optimizations can be performed. For instance, slots classes take up less memory, as can be measured using Pympler.

03:04 In this particular example, the slot class is about 20% faster. So now that you’ve learned a great deal about data classes, let’s sum up everything that you’ve covered.

mp on April 28, 2023

I hadn’t heard of slots before. I tried messing around with this afterwards, I found that I couldn’t set defaults when slots had been defined. But then further messing around, it seems that now you don’t even have to declare slots manually (at least if all fields are to use slots):

@dataclass(frozen=True, slots=True)
class Position:
    name: str
    longitude: float = 0.0
    latitude: float = 0.0

pos = Position("oslo", 5, 10)

>>> pos.__slots__
('name', 'longitude', 'latitude')

Looks like the slots parameter was added in 3.10 (this course was earlier, using 3.9). docs.python.org/3/library/dataclasses.html

Become a Member to join the conversation.