# DevInfra CI Directory

> **Warning** This folder is still under construction. It is part of an ongoing
> effort to improve the structure of CI and build related files within the
> TensorFlow repo. This warning will be removed when the contents of this
> directory are stable and appropriate documentation around its usage is in
> place.

Maintainer: TensorFlow DevInfra

Issue Reporting: File an issue against this repo and tag

   *
   * @deprecated Prefer {@linkplain ExecutionError(Error)} a constructor that accepts a cause: Users
   *     of this class typically expect for instances to have a non-null cause. At the moment, you
   *     can <i>usually</i> still preserve behavior by passing an explicit {@code null} cause. Note,
   *     however, that passing an explicit {@code null} cause prevents anyone from calling {@link

        assertSame(b, reused, "Buffer should be reused from cache");
        // And it should still be zeroed
        for (byte value : reused) {
            assertEquals(0, value, "Reused buffer must be zeroed");
        }
    }

    // Edge: cache size 0 means nothing is cached; release still zeroes argument
    @Test
    @DisplayName("Cache size 0: release zeroes but does not cache; getBuffer allocates new")

But this example is still valid and it shows how to interact with the internal components.

///

We can also use this same approach to access the request body in an exception handler.

The next sections are **not necessarily "advanced"**.

And it's possible that for your use case, the solution is in one of them.

///

## Read the Tutorial first { #read-the-tutorial-first }

You could still use most of the features in **FastAPI** with the knowledge from the main [Tutorial - User Guide](../tutorial/index.md){.internal-link target=_blank}.

       * Try adding the CUDA, CUPTI, and cuDNN installation directories to the $LD_LIBRARY_PATH environment variable.
       * Refer [linux setup guide](https://www.tensorflow.org/install/gpu#linux_setup).
     * If error still persists then, apparently your CPU model does not support AVX instruction sets.
       * Refer [hardware requirements](https://www.tensorflow.org/install/pip#hardware-requirements).
   

    """
    Regression test ensuring that when a method imported from another module
    is decorated with something that sets the __wrapped__ attribute (functools.wraps),
    then the types are still processed correctly, including dereferencing of forward
    references.
    """
    app = FastAPI()
    client = TestClient(app)
    app.post("/endpoint")(passthrough(forwardref_method))

It had the best performance benchmarks at the time (only surpassed by Starlette).

At first, it didn't have an automatic API documentation web UI, but I knew I could add Swagger UI to it.

It had a dependency injection system. It required pre-registration of components, as other tools discussed above. But still, it was a great feature.

# TensorFlow continuous integration

> **Warning** This folder is still under construction. It is part of an ongoing
> effort to improve the structure of CI and build related files within the
> TensorFlow repo. This warning will be removed when the contents of this
> directory are stable and appropriate documentation around its usage is in
> place.

Maintainer: TensorFlow DevInfra

   *
   * @deprecated Prefer {@linkplain ExecutionError(Error)} a constructor that accepts a cause: Users
   *     of this class typically expect for instances to have a non-null cause. At the moment, you
   *     can <i>usually</i> still preserve behavior by passing an explicit {@code null} cause. Note,
   *     however, that passing an explicit {@code null} cause prevents anyone from calling {@link