Goal Type will default to maximize.

Maximize the goal metric.

Minimize the goal metric.

The job has been cancelled.

The job is being cancelled.

The job failed.

The service is preparing to run the job.

The job has been just created and processing has not yet begun.

The job is in progress.

The job state is unspecified.

The job completed successfully.

An operation to create a new version.

An operation to delete an existing model.

An operation to delete an existing version.

Unspecified operation type.

The parameter is categorical, with a value chosen from the categories field.

The parameter is real valued, with a fixed set of feasible points.

Type for real-valued parameters.

Type for integral parameters.

By default, no scaling is applied.

You must specify a valid type.

Scales the feasible space to (0, 1) linearly.

Scales the feasible space logarithmically to (0, 1).

Scales the feasible space "reverse" logarithmically to (0, 1).

Unspecified format.

The source file is a text file with instances separated by the new-line character.

The source file is a TFRecord file.

The source file is a GZIP-compressed TFRecord file.

A single worker instance.

A single worker instance [with a GPU](ml/docs/how-tos/using-gpus).

The CUSTOM tier is not a set tier, but rather enables you to use your own cluster specification.

A large number of workers with many parameter servers.

Many workers and a few parameter servers.

Request message for the CancelJob method.

Request message for the CreateJob method.

Request message for the CreateModel method.

Uploads the provided trained model version to Cloud Machine Learning.

Request message for the DeleteModel method.

Request message for the DeleteVerionRequest method.

Requests service account information associated with a project.

Returns service account information associated with a project.

Request message for the GetJob method.

Request message for the GetModel method.

Request message for the GetVersion method.

Represents the result of a single hyperparameter tuning trial from a training job.

An observed value of a metric.

Represents a set of hyperparameters to optimize.

Represents a training or prediction job.

Request message for the ListJobs method.

Response message for the ListJobs method.

Request message for the ListModels method.

Response message for the ListModels method.

Request message for the ListVersions method.

Response message for the ListVersions method.

Options for manually scaling a model.

Represents a machine learning solution.

Represents the metadata of the long-running operation.

Represents a single hyperparameter to optimize.

Represents input parameters for a prediction job.

Represents results of a prediction job.

Request for predictions to be issued against a trained model.

Request message for the SetDefaultVersion request.

Represents input parameters for a training job.

Represents results of a training job.

Represents a version of the model.

The available types of optimization goals.

Describes the job state.

The operation type.

The type of the parameter.

The type of scaling that should be applied to this parameter.

The format used to separate data instances in the source files.

A scale tier is an abstract representation of the resources Cloud ML will allocate to a training job.