Arg<TF_NumberTensor, [{Must be a scalar.}]>:$beta1_power,
    Arg<TF_NumberTensor, [{Must be a scalar.}]>:$beta2_power,
    Arg<TF_NumberTensor, [{Scaling factor. Must be a scalar.}]>:$lr,
    Arg<TF_NumberTensor, [{Momentum factor. Must be a scalar.}]>:$beta1,
    Arg<TF_NumberTensor, [{Momentum factor. Must be a scalar.}]>:$beta2,
    Arg<TF_NumberTensor, [{Ridge term. Must be a scalar.}]>:$epsilon,

        Keras training loops like `fit`/`evaluate`, the unreduced vector loss is
        passed to the optimizer but the reported loss will be a scalar value.
    *   `SUM`: Scalar sum of weighted losses. 4. `SUM_OVER_BATCH_SIZE`: Scalar
        `SUM` divided by number of elements in losses. This reduction type is
        not supported when used with `tf.distribute.Strategy` outside of

            "format": "int32",
            "type": "integer"
          }
        },
        "type": "object"
      },
      "io.k8s.api.apps.v1.StatefulSet": {