private fun KtDiagnosticCheckerFilter.asLLFilter() = when (this) {
        KtDiagnosticCheckerFilter.ONLY_COMMON_CHECKERS -> DiagnosticCheckerFilter.ONLY_COMMON_CHECKERS
        KtDiagnosticCheckerFilter.ONLY_EXTENDED_CHECKERS -> DiagnosticCheckerFilter.ONLY_EXTENDED_CHECKERS
        KtDiagnosticCheckerFilter.EXTENDED_AND_COMMON_CHECKERS -> DiagnosticCheckerFilter.EXTENDED_AND_COMMON_CHECKERS
    }

-   x86 and arm64
-   CPU-only, or with NVIDIA CUDA support (Linux only), or with TPUs

## How to Test Your Changes to TensorFlow

You may check how your changes will affect TensorFlow by:

1. Creating a PR and observing the presubmit test results
2. Running the CI scripts locally, as explained below
3. **Google employees only**: Google employees can use an internal-only tool

    go mod tidy
    go mod vendor

Use caution when passing '-u' to 'go get'. The '-u' flag updates
modules providing all transitively imported packages, not only
the module providing the target package.

Note that 'go mod vendor' only copies packages that are transitively
imported by packages in the current module. If a new package is needed,

      // it is a name, all IPs for that name should be tried. If it is an IP address, only that IP
      // address should be tried.
      val address = address ?: return hostName

      // The InetSocketAddress has a specific address: we should only try that address. Therefore we
      // return the address and ignore any host name that may be available.
      return address.hostAddress

[Using boto3](https://github.com/minio/minio/blob/master/docs/extensions/s3zip/examples/boto3/main.py)

## Requirements and limits

- ListObjectsV2 can only list the most recent ZIP archive version of your object, applicable only for versioned buckets.
- ListObjectsV2 API calls must be used to list zip file content.
- Range requests for GetObject/HeadObject for individual files from zip is not supported.

document also outlines what issues we consider as TensorFlow security
vulnerabilities.

We recognize issues as vulnerabilities only when they occur in scenarios that we
outline as safe; issues that have a security impact only when TensorFlow is used
in a discouraged way (e.g. running untrusted models or checkpoints, data parsing
outside of the safe formats, etc.) are not treated as vulnerabilities..

jcifs-ng will be a proper choice for many users. 
There are a lot of SMB devices in the world.
Some of them only work with the old jcifs library.
If you want to support many SMB devices, CodeLibs jcifs library will be helpful.
For example, since [Fess](https://github.com/codelibs/fess) needs to support many SMB devices, it uses this library.

          "\r\n".encodeUtf8(),
          // 1.  "--"    No more parts.
          "--".encodeUtf8(),
          // 2.  " "     Optional whitespace. Only used if there are more parts.
          " ".encodeUtf8(),
          // 3.  "\t"    Optional whitespace. Only used if there are more parts.
          "\t".encodeUtf8(),
        )
    }

For example, if you need to store it in a database.

For that, **FastAPI** provides a `jsonable_encoder()` function.

## Using the `jsonable_encoder`

Let's imagine that you have a database `fake_db` that only receives JSON compatible data.

For example, it doesn't receive `datetime` objects, as those are not compatible with JSON.

    applyDefaultSettings(os = os, arch = arch, buildJvm = BuildToolBuildJvm, timeout = 180)

    val testsWithOs = model.stages.filter {
        it.stageName in listOf(
            StageName.QUICK_FEEDBACK_LINUX_ONLY,
            StageName.QUICK_FEEDBACK,
            StageName.PULL_REQUEST_FEEDBACK,
            StageName.READY_FOR_NIGHTLY,
        )
    }.flatMap { it.functionalTests }.filter { it.os == os }