Python Dataclasses

What Is A Dataclass?

A dataclass is a class designed mainly to hold data. The dataclasses module can generate common methods such as __init__(), __repr__(), and equality checks.

from dataclasses import dataclass

@dataclass
class Student:
    name: str
    score: int

Now you can create objects without writing an initializer by hand.

student = Student("Asha", 88)
print(student)

Defaults

You can give fields default values.

@dataclass
class Task:
    title: str
    done: bool = False

Fields without defaults must come before fields with defaults.

Mutable Defaults

Avoid using mutable values directly as defaults.

from dataclasses import field

@dataclass
class Team:
    members: list[str] = field(default_factory=list)

default_factory creates a new list for each instance.

Frozen Dataclasses

Use frozen=True to make instances mostly immutable.

@dataclass(frozen=True)
class Point:
    x: int
    y: int

Trying to assign to point.x later raises an error.

Post Initialization

Use __post_init__() for validation or derived fields after the generated __init__() runs.

@dataclass
class Score:
    value: int

    def __post_init__(self):
        if self.value < 0:
            raise ValueError("score cannot be negative")

Common Mistakes

A common mistake is using dataclasses for behavior-heavy objects where a normal class would communicate intent better.

Another mistake is forgetting default_factory for lists, dictionaries, and sets. Shared mutable defaults create surprising bugs.

Quick Summary

• @dataclass generates common class methods.
• Type annotations define dataclass fields.
• Defaults are allowed after required fields.
• Use field(default_factory=...) for mutable defaults.
• frozen=True makes instances harder to mutate.
• __post_init__() supports validation after initialization.

Practice Quiz

Now practice this topic with MCQs and explanations:

Start the Python Dataclasses quiz