![Final graph](g3doc/images/final_graph.png)

## TAC components

### Hardwares

Hardwares are used to modeling target device capabilities & also ops cost for
the target devices.

We have already modeled `cpu_hardware` & `gpu_hardware` as well as the
`nnapi_hardware`.

### Passes

#### Target Annotation Pass
In this pass, every op will be targeted with the user specified targets based on

	check.singleValue(x)
}

// exclude reports an error if x.mode is in modeset and sets x.mode to invalid.
// The modeset may contain any of 1<<novalue, 1<<builtin, 1<<typexpr.
func (check *Checker) exclude(x *operand, modeset uint) {
	if modeset&(1<<x.mode) != 0 {
		var msg string
		var code Code
		switch x.mode {
		case novalue:
			if modeset&(1<<typexpr) != 0 {
				msg = "%s used as value"
			} else {

    disabled = user_dict["disabled"],
    hashed_password = user_dict["hashed_password"],
)
```

!!! info
    Eine ausführlichere Erklärung von `**user_dict` finden Sie in [der Dokumentation für **Extra Modelle**](../extra-models.md#uber-user_indict){.internal-link target=_blank}.

## Den Token zurückgeben

Die Response des `token`-Endpunkts muss ein JSON-Objekt sein.

```Python hl_lines="6"
{!../../../docs_src/python_types/tutorial010.py!}
```

Et vous aurez accès, encore une fois, au support complet offert par l'éditeur :

<img src="/img/python-types/image06.png">

## Les modèles Pydantic

<a href="https://docs.pydantic.dev/" class="external-link" target="_blank">Pydantic</a> est une bibliothèque Python pour effectuer de la validation de données.

  tfr.return %ret : !tfr.tensor
}

//=================>  signatures of the primitive ops with kernels, modeled as external TFR function <==
tfr.func @tf__risc_cast_(!tfr.tensor, !tfr.attr{tfr.name="Tout"}) -> !tfr.tensor<Tout> attributes{Tout}
tfr.func @tf__risc_add_(!tfr.tensor<T>, !tfr.tensor<T>) -> !tfr.tensor<T> attributes{T}

.Customizing the XML
====
include::sample[dir="snippets/ide/idea/kotlin",files="build.gradle.kts[tags=project-with-xml]"]

		}
	}
	return false
}

func (boss *ImportBoss) Verify(pkg *packages.Package) []error {
	pkgDir := packageDir(pkg)
	if pkgDir == "" {
		// This Package has no usable files, e.g. only tests, which are modelled in
		// a distinct Package.
		return nil
	}

	restrictionFiles, err := recursiveRead(filepath.Join(pkgDir, rulesFileName))
	if err != nil {
		return []error{fmt.Errorf("error finding rules file: %w", err)}

.Customizing the XML
====
include::sample[dir="snippets/ide/eclipse/kotlin",files="build.gradle.kts[tags=wtp-with-xml]"]

   * example is when a server has just booted, and it is mostly busy with getting itself up to
   * speed).
   *
   * To deal with such scenarios, we add an extra dimension, that of "past underutilization",
   * modeled by "storedPermits" variable. This variable is zero when there is no underutilization,
   * and it can grow up to maxStoredPermits, for sufficiently large underutilization. So, the

<img src="/img/python-types/image06.png">

Beachten Sie, das bedeutet: „`one_person` ist eine **Instanz** der Klasse `Person`“.

Es bedeutet nicht: „`one_person` ist die **Klasse** genannt `Person`“.

## Pydantic Modelle

<a href="https://pydantic-docs.helpmanual.io/" class="external-link" target="_blank">Pydantic</a> ist eine Python-Bibliothek für die Validierung von Daten.

Sie deklarieren die „Form“ der Daten als Klassen mit Attributen.