used to shard the checkpoint with a maximum shard file size. Users with
      advanced use cases can also write their own custom
      `tf.train.experimental.ShardingCallback`s.

* `tf.train.CheckpointOptions`
    * Added `experimental_skip_slot_variables` (a boolean option) to skip
    restoring of optimizer slot variables in a checkpoint.

*   `tf.saved_model.SaveOptions`

/// note | Détails techniques

ASGI est une nouvelle "norme" développée par les membres de l'équipe principale de Django. Il ne s'agit pas encore d'une "norme Python" (un PEP), bien qu'ils soient en train de le faire.

            @Nullable Throwable error) {}

    /**
     * If this is an accumulating log, it will "drain" this instance. It returns the accumulated log entries, and
     * also "resets" this instance to empty (initial) state.
     */
    @Nonnull
    default List<Entry> drain() {
        return List.of();
    }

# TODO(b/173549186): Move Google-internal TF code out of learning/brain
package_group(
    name = "internal",
    packages = [
        "//devtools/python/indexer/...",
        "//learning/brain/keras/...",
        "//learning/brain/mlir/...",
        "//learning/brain/tfrt/...",
        "//learning/lib/ami/simple_ml/...",
        "//learning/pathways/...",

	// ReplicateQueued - replication being queued trail
	ReplicateQueued = "replicate:queue"

	// ReplicateExisting - audit trail for existing objects replication
	ReplicateExisting = "replicate:existing"
	// ReplicateExistingDelete - audit trail for delete replication triggered for existing delete markers
	ReplicateExistingDelete = "replicate:existing:delete"

    * If you know Python types you know how to use Pydantic.
* Plays nicely with your **<abbr title="Integrated Development Environment: similar to a code editor">IDE</abbr>/<abbr title="A program that checks for code errors">linter</abbr>/brain**:
    * Because pydantic data structures are just instances of classes you define; auto-completion, linting, mypy and your intuition should all work properly with your validated data.
* Validate **complex structures**:

* Kullandığın geliştirme araçları ile iyi çalışır **<abbr title="Integrated Development Environment, kod editörüne benzer">IDE</abbr>/<abbr title="Code errorlarınızı inceleyen program">linter</abbr>/brain**:
    * Pydantic'in veri yapıları aslında sadece senin tanımladığın classlar; Bu yüzden doğrulanmış dataların ile otomatik tamamlama, linting ve mypy'ı kullanarak sorunsuz bir şekilde çalışabilirsin

            // in case of mvnsh the context.logger != context.invokerRequest.parserRequest.logger
            List<Logger.Entry> entries =
                    context.invokerRequest.parserRequest().logger().drain();
            printErrors(
                    context,
                    context.invokerRequest
                            .parserRequest()
                            .args()

# Tests

tf_cuda_library(
    name = "c_test_util",
    testonly = 1,
    srcs = ["c_test_util.cc"],
    hdrs = ["c_test_util.h"],
    visibility = [
        "//learning/brain:__subpackages__",
        "//tensorflow:__subpackages__",
    ],
    deps = [
        ":c_api",
        ":c_api_experimental",
        "//tensorflow/core:lib",
        "//tensorflow/core:protos_all_cc",

* **更简单**：
    * 没有新的模式定义 micro-language 需要学习。
    * 如果你知道 Python types，你就知道如何使用 Pydantic。
* 和你 **<abbr title="集成开发环境，和代码编辑器类似">IDE</abbr>/<abbr title="一个检查代码错误的程序">linter</abbr>/brain** 适配:
    * 因为 pydantic 数据结构仅仅是你定义的类的实例；自动补全，linting，mypy 以及你的直觉应该可以和你验证的数据一起正常工作。
* 验证**复杂结构**:
    * 使用分层的 Pydantic 模型, Python `typing`的 `List` 和 `Dict` 等等。
    * 验证器使我们能够简单清楚的将复杂的数据模式定义、检查并记录为 JSON Schema。