```Python hl_lines="14"
{!../../../docs_src/body_nested_models/tutorial001.py!}
```

이는 `tags`를 항목 리스트로 만듭니다. 각 항목의 타입을 선언하지 않더라도요.

## 타입 매개변수가 있는 리스트 필드

하지만 파이썬은 내부의 타입이나 "타입 매개변수"를 선언할 수 있는 특정 방법이 있습니다:

### typing의 `List` 임포트

먼저, 파이썬 표준 `typing` 모듈에서 `List`를 임포트합니다:

```Python hl_lines="1"
{!../../../docs_src/body_nested_models/tutorial002.py!}
```

### 타입 매개변수로 `List` 선언

## Modelos aninhados

Cada atributo de um modelo Pydantic tem um tipo.

Mas esse tipo pode ser outro modelo Pydantic.

Portanto, você pode declarar "objects" JSON profundamente aninhados com nomes, tipos e validações de atributos específicos.

Tudo isso, aninhado arbitrariamente.

### Defina um sub-modelo

Por exemplo, nós podemos definir um modelo `Image`:

```Python hl_lines="9-11"

=== "Python 3.10+"

    ```Python hl_lines="12"
    {!> ../../../docs_src/body_nested_models/tutorial002_py310.py!}
    ```

=== "Python 3.9+"

    ```Python hl_lines="14"
    {!> ../../../docs_src/body_nested_models/tutorial002_py39.py!}
    ```

=== "Python 3.8+"

    ```Python hl_lines="14"
    {!> ../../../docs_src/body_nested_models/tutorial002.py!}
    ```

## Set 类型

但是随后我们考虑了一下，意识到标签不应该重复，它们很大可能会是唯一的字符串。

    ```Python hl_lines="14"
    {!> ../../../docs_src/body_nested_models/tutorial002.py!}
    ```

=== "🐍 3️⃣.9️⃣ & 🔛"

    ```Python hl_lines="14"
    {!> ../../../docs_src/body_nested_models/tutorial002_py39.py!}
    ```

=== "🐍 3️⃣.1️⃣0️⃣ & 🔛"

    ```Python hl_lines="12"
    {!> ../../../docs_src/body_nested_models/tutorial002_py310.py!}
    ```

## ⚒ 🆎

```Python hl_lines="12"
{!../../../docs_src/body_nested_models/tutorial001.py!}
```

これにより、各項目の型は宣言されていませんが、`tags`はある項目のリストになります。

## タイプパラメータを持つリストのフィールド

しかし、Pythonには型や「タイプパラメータ」を使ってリストを宣言する方法があります:

### typingの`List`をインポート

まず、Pythonの標準の`typing`モジュールから`List`をインポートします:

```Python hl_lines="1"
{!../../../docs_src/body_nested_models/tutorial002.py!}
```

### タイプパラメータを持つ`List`の宣言

from fastapi import FastAPI
from fastapi.testclient import TestClient
from pydantic import BaseModel

app = FastAPI()


class Model(BaseModel):
    pass


class Model2(BaseModel):
    a: Model


class Model3(BaseModel):
    c: Model
    d: Model2


@app.get("/", response_model=Model3)
def f():
    return {"c": {}, "d": {"a": {}}}


client = TestClient(app)


def test_get_api_route():

TensorFlow
[**models**](https://developers.google.com/machine-learning/glossary/#model) (to
use a term commonly used by machine learning practitioners) are expressed as
programs that TensorFlow executes. TensorFlow programs are encoded as
computation
[**graphs**](https://developers.google.com/machine-learning/glossary/#graph).
Since models are practically programs that TensorFlow executes, using untrusted

from sqlalchemy.orm import Session

from . import models, schemas


def get_user(db: Session, user_id: int):
    return db.query(models.User).filter(models.User.id == user_id).first()


def get_user_by_email(db: Session, email: str):
    return db.query(models.User).filter(models.User.email == email).first()


def get_users(db: Session, skip: int = 0, limit: int = 100):
    return db.query(models.User).offset(skip).limit(limit).all()

from sqlalchemy.orm import Session

from . import models, schemas


def get_user(db: Session, user_id: int):
    return db.query(models.User).filter(models.User.id == user_id).first()


def get_user_by_email(db: Session, email: str):
    return db.query(models.User).filter(models.User.email == email).first()


def get_users(db: Session, skip: int = 0, limit: int = 100):
    return db.query(models.User).offset(skip).limit(limit).all()

from fastapi import FastAPI


def fake_answer_to_everything_ml_model(x: float):
    return x * 42


ml_models = {}


@asynccontextmanager
async def lifespan(app: FastAPI):
    # Load the ML model
    ml_models["answer_to_everything"] = fake_answer_to_everything_ml_model
    yield
    # Clean up the ML models and release the resources
    ml_models.clear()


app = FastAPI(lifespan=lifespan)