Let's imagine that you have some **machine learning models** that you want to use to handle requests. 🤖

The same models are shared among requests, so, it's not one model per request, or one per user or something similar.

Let's imagine that loading the model can **take quite some time**, because it has to read a lot of **data from disk**. So you don't want to do it for every request.

          <description>Specify the version when the Mojo was added to the API. Similar to Javadoc since.</description>
        </field>
        <field>
          <name>deprecated</name>
          <version>1.0.0+</version>
          <type>String</type>
          <description>
            Description with the reason of Mojo deprecation. Similar to Javadoc {@code @deprecated}

 * object.
 *
 * <p><b>Comparison to {@code java.util.Optional} (JDK 8 and higher):</b> A new {@code Optional}
 * class was added for Java 8. The two classes are extremely similar, but incompatible (they cannot
 * share a common supertype). <i>All</i> known differences are listed either here or with the
 * relevant methods below.
 *
 * <ul>

It achieves something somewhat similar to what can be done with Flask-apispec.

It has an integrated dependency injection system, inspired by Angular 2. It requires pre-registering the "injectables" (like all the other dependency injection systems I know), so, it adds to the verbosity and code repetition.

As the parameters are described with TypeScript types (similar to Python type hints), editor support is quite good.

 * object.
 *
 * <p><b>Comparison to {@code java.util.Optional} (JDK 8 and higher):</b> A new {@code Optional}
 * class was added for Java 8. The two classes are extremely similar, but incompatible (they cannot
 * share a common supertype). <i>All</i> known differences are listed either here or with the
 * relevant methods below.
 *
 * <ul>

Ele alcança de uma forma similar ao que pode ser feito com o Flask-apispec.

Ele tem um sistema de injeção de dependência integrado, inspirado pelo Angular dois. É necessário fazer o pré-registro dos "injetáveis" (como todos os sistemas de injeção de dependência que conheço), então, adicionando verbosidade e repetição de código.

    for (int i = 0; i < grad_outputs.size(); i++) {
      auto grad_input = grad_outputs[i];
      // TODO(srbs): Should we add a copy contructor to AbstractTensorHandle
      // that takes care of this similar to `Tensor`?
      if (grad_input) {
        grad_input->Ref();
      }
      grad_inputs[i] = grad_input;
    }
    return absl::OkStatus();
  }
  ~IdentityNGradientFunction() override {}
};

## Containers and Processes

A **container image** normally includes in its metadata the default program or command that should be run when the **container** is started and the parameters to be passed to that program. Very similar to what would be if it was in the command line.

When a **container** is started, it will run that command/program (although you can override it and make it run a different command/program).

  optional int32 targetAverageUtilization = 2;

  // targetAverageValue is the target value of the average of the
  // resource metric across all relevant pods, as a raw value (instead of as
  // a percentage of the request), similar to the "pods" metric source type.
  // +optional
  optional k8s.io.apimachinery.pkg.api.resource.Quantity targetAverageValue = 3;

  // container is the name of the container in the pods of the scaling target

### Always `raise` in Dependencies with `yield` and `except`

If you catch an exception in a dependency with `yield`, unless you are raising another `HTTPException` or similar, you should re-raise the original exception.