`tf.linalg.matvec` convenience function.
    *   `tf.einsum()`raises `ValueError` for unsupported equations like
        `"ii->"`.
    *   Add DCT-I and IDCT-I in `tf.signal.dct` and `tf.signal.idct`.
    *   Add LU decomposition op.
    *   Add quantile loss to gradient boosted trees in estimator.
    *   Add `round_mode` to `QuantizeAndDequantizeV2` op to select rounding
        algorithm.

В Pydantic v1 метод назывался `.dict()`, в Pydantic v2 он помечен как устаревший (но всё ещё поддерживается) и переименован в `.model_dump()`.

В примерах здесь используется `.dict()` для совместимости с Pydantic v1, но если вы используете Pydantic v2, следует использовать `.model_dump()`.

///

### Про `**user_in.dict()` { #about-user-in-dict }

#### `.dict()` из Pydantic { #pydantics-dict }

No Pydantic v1 o método se chamava `.dict()`, ele foi descontinuado (mas ainda é suportado) no Pydantic v2 e renomeado para `.model_dump()`.

Os exemplos aqui usam `.dict()` por compatibilidade com o Pydantic v1, mas você deve usar `.model_dump()` se puder usar o Pydantic v2.

///

### Sobre `**user_in.dict()` { #about-user-in-dict }

#### O `.dict()` do Pydantic { #pydantics-dict }

    schemas: Optional[dict[str, Union[Schema, Reference]]] = None
    responses: Optional[dict[str, Union[Response, Reference]]] = None
    parameters: Optional[dict[str, Union[Parameter, Reference]]] = None
    examples: Optional[dict[str, Union[Example, Reference]]] = None
    requestBodies: Optional[dict[str, Union[RequestBody, Reference]]] = None
    headers: Optional[dict[str, Union[Header, Reference]]] = None

    items: list[dict[str, Any]]
    metadata: dict[str, Any]


class LargeOut(BaseModel):
    items: list[dict[str, Any]]
    metadata: dict[str, Any]


app = FastAPI()


@app.post("/sync/validated", response_model=ItemOut)
def sync_validated(item: ItemIn, dep: Annotated[int, Depends(dep_b)]):
    return ItemOut(name=item.name, value=item.value, dep=dep)


@app.get("/sync/dict-no-response-model")

@app.get("/by-alias/dict", response_model=Model)
def by_alias_dict():
    return {"alias": "Foo"}


@app.get("/by-alias/model", response_model=Model)
def by_alias_model():
    return Model(alias="Foo")


@app.get("/by-alias/list", response_model=list[Model])
def by_alias_list():
    return [{"alias": "Foo"}, {"alias": "Bar"}]


@app.get("/no-alias/dict", response_model=ModelNoAlias)

def _remap_definitions_and_field_mappings(
    *,
    model_name_map: ModelNameMap,
    definitions: dict[str, Any],
    field_mapping: dict[
        tuple[ModelField, Literal["validation", "serialization"]], JsonSchemaValue
    ],
) -> tuple[
    dict[tuple[ModelField, Literal["validation", "serialization"]], JsonSchemaValue],
    dict[str, Any],
]:
    old_name_to_new_name_map = {}
    for field_key, schema in field_mapping.items():

    operation_ids: set[str],
    model_name_map: ModelNameMap,
    field_mapping: dict[
        tuple[ModelField, Literal["validation", "serialization"]], dict[str, Any]
    ],
    separate_input_output_schemas: bool = True,
) -> tuple[dict[str, Any], dict[str, Any], dict[str, Any]]:
    path = {}
    security_schemes: dict[str, Any] = {}
    definitions: dict[str, Any] = {}
    assert route.methods is not None, "Methods must be a list"

    pass


class OtherDependencyError(Exception):
    pass


async def asyncgen_state(state: dict[str, str] = Depends(get_state)):
    state["/async"] = "asyncgen started"
    yield state["/async"]
    state["/async"] = "asyncgen completed"


def generator_state(state: dict[str, str] = Depends(get_state)):
    state["/sync"] = "generator started"
    yield state["/sync"]

En Pydantic v1 el método se llamaba `.dict()`, fue deprecado (pero aún soportado) en Pydantic v2, y renombrado a `.model_dump()`.

Los ejemplos aquí usan `.dict()` para compatibilidad con Pydantic v1, pero deberías usar `.model_dump()` en su lugar si puedes usar Pydantic v2.

///

### Acerca de `**user_in.dict()` { #about-user-in-dict }

#### `.dict()` de Pydantic { #pydantics-dict }