from the `TopK` op in its handling of ties; if multiple classes have the
same prediction value and straddle the top-`k` boundary, all of those
classes are considered to be in the top `k`.

More formally, let

  \\(predictions_i\\) be the predictions for all classes for example `i`,
  \\(targets_i\\) be the target class for example `i`,

        not supported when used with `tf.distribute.Strategy` outside of
        built-in training loops like `tf.keras` `compile`/`fit`.
    *   Wraps losses passed to the `compile` API (strings and v1 losses) which
        are not instances of v2 `Loss` class in `LossWrapper` class. => All
        losses will now use `SUM_OVER_BATCH_SIZE` reduction as default.

                            of the authorized client certificates.
                          items:
                            type: string
                          type: array
                        verifyCertificateSpki:
                          description: An optional list of base64-encoded SHA-256
                            hashes of the SPKIs of authorized client certificates.

                            of the authorized client certificates.
                          items:
                            type: string
                          type: array
                        verifyCertificateSpki:
                          description: An optional list of base64-encoded SHA-256
                            hashes of the SPKIs of authorized client certificates.

            "type": "string"
          },
          "whenScaled": {

        "properties": {
          "causes": {
            "description": "The Causes array includes more details associated with the StatusReason failure. Not all StatusReasons may provide detailed causes.",
            "items": {
              "allOf": [
                {