Docs Home Pulumi IaC Pulumi ESC Pulumi Cloud Packages Tutorials
  1. Packages
  2. Google Cloud Native
  3. API Docs
  4. ml
  5. ml/v1
  6. Job

Google Cloud Native is in preview. Google Cloud Classic is fully supported.

Google Cloud Native v0.32.0 published on Wednesday, Nov 29, 2023 by Pulumi
pulumi/pulumi-google-native

google-native.ml/v1.Job

Explore with Pulumi AI

google-native logo

Google Cloud Native is in preview. Google Cloud Classic is fully supported.

Google Cloud Native v0.32.0 published on Wednesday, Nov 29, 2023 by Pulumi
pulumi/pulumi-google-native

    Creates a training or a batch prediction job. Auto-naming is currently not supported for this resource. Note - this resource’s API doesn’t support deletion. When deleted, the resource will persist on Google Cloud even though it will be deleted from Pulumi state.

    Create Job Resource

    Resources are created with functions called constructors. To learn more about declaring and configuring resources, see Resources.

    Constructor syntax

    new Job(name: string, args: JobArgs, opts?: CustomResourceOptions);
    @overload
def Job(resource_name: str,
        args: JobArgs,
        opts: Optional[ResourceOptions] = None)

@overload
def Job(resource_name: str,
        opts: Optional[ResourceOptions] = None,
        job_id: Optional[str] = None,
        etag: Optional[str] = None,
        labels: Optional[Mapping[str, str]] = None,
        prediction_input: Optional[GoogleCloudMlV1__PredictionInputArgs] = None,
        prediction_output: Optional[GoogleCloudMlV1__PredictionOutputArgs] = None,
        project: Optional[str] = None,
        training_input: Optional[GoogleCloudMlV1__TrainingInputArgs] = None,
        training_output: Optional[GoogleCloudMlV1__TrainingOutputArgs] = None)
    func NewJob(ctx *Context, name string, args JobArgs, opts ...ResourceOption) (*Job, error)
    public Job(string name, JobArgs args, CustomResourceOptions? opts = null)
    public Job(String name, JobArgs args)
public Job(String name, JobArgs args, CustomResourceOptions options)

    type: google-native:ml/v1:Job
properties: # The arguments to resource properties.
options: # Bag of options to control resource's behavior.

    Parameters

    name string
    The unique name of the resource.
    args JobArgs
    The arguments to resource properties.
    opts CustomResourceOptions
    Bag of options to control resource's behavior.
    resource_name str
    The unique name of the resource.
    args JobArgs
    The arguments to resource properties.
    opts ResourceOptions
    Bag of options to control resource's behavior.
    ctx Context
    Context object for the current deployment.
    name string
    The unique name of the resource.
    args JobArgs
    The arguments to resource properties.
    opts ResourceOption
    Bag of options to control resource's behavior.
    name string
    The unique name of the resource.
    args JobArgs
    The arguments to resource properties.
    opts CustomResourceOptions
    Bag of options to control resource's behavior.
    name String
    The unique name of the resource.
    args JobArgs
    The arguments to resource properties.
    options CustomResourceOptions
    Bag of options to control resource's behavior.

    Constructor example

    The following reference example uses placeholder values for all input properties.

    var examplejobResourceResourceFromMlv1 = new GoogleNative.Ml.V1.Job("examplejobResourceResourceFromMlv1", new()
{
    JobId = "string",
    Etag = "string",
    Labels = 
    {
        { "string", "string" },
    },
    PredictionInput = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__PredictionInputArgs
    {
        DataFormat = GoogleNative.Ml.V1.GoogleCloudMlV1__PredictionInputDataFormat.DataFormatUnspecified,
        InputPaths = new[]
        {
            "string",
        },
        OutputPath = "string",
        Region = "string",
        BatchSize = "string",
        MaxWorkerCount = "string",
        ModelName = "string",
        OutputDataFormat = GoogleNative.Ml.V1.GoogleCloudMlV1__PredictionInputOutputDataFormat.DataFormatUnspecified,
        RuntimeVersion = "string",
        SignatureName = "string",
        Uri = "string",
        VersionName = "string",
    },
    PredictionOutput = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__PredictionOutputArgs
    {
        ErrorCount = "string",
        NodeHours = 0,
        OutputPath = "string",
        PredictionCount = "string",
    },
    Project = "string",
    TrainingInput = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__TrainingInputArgs
    {
        PackageUris = new[]
        {
            "string",
        },
        ScaleTier = GoogleNative.Ml.V1.GoogleCloudMlV1__TrainingInputScaleTier.Basic,
        Region = "string",
        PythonModule = "string",
        ParameterServerConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfigArgs
        {
            AcceleratorConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__AcceleratorConfigArgs
            {
                Count = "string",
                Type = GoogleNative.Ml.V1.GoogleCloudMlV1__AcceleratorConfigType.AcceleratorTypeUnspecified,
            },
            ContainerArgs = new[]
            {
                "string",
            },
            ContainerCommand = new[]
            {
                "string",
            },
            DiskConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__DiskConfigArgs
            {
                BootDiskSizeGb = 0,
                BootDiskType = "string",
            },
            ImageUri = "string",
            TpuTfVersion = "string",
        },
        EvaluatorConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfigArgs
        {
            AcceleratorConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__AcceleratorConfigArgs
            {
                Count = "string",
                Type = GoogleNative.Ml.V1.GoogleCloudMlV1__AcceleratorConfigType.AcceleratorTypeUnspecified,
            },
            ContainerArgs = new[]
            {
                "string",
            },
            ContainerCommand = new[]
            {
                "string",
            },
            DiskConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__DiskConfigArgs
            {
                BootDiskSizeGb = 0,
                BootDiskType = "string",
            },
            ImageUri = "string",
            TpuTfVersion = "string",
        },
        Hyperparameters = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__HyperparameterSpecArgs
        {
            Goal = GoogleNative.Ml.V1.GoogleCloudMlV1__HyperparameterSpecGoal.GoalTypeUnspecified,
            Params = new[]
            {
                new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ParameterSpecArgs
                {
                    ParameterName = "string",
                    Type = GoogleNative.Ml.V1.GoogleCloudMlV1__ParameterSpecType.ParameterTypeUnspecified,
                    CategoricalValues = new[]
                    {
                        "string",
                    },
                    DiscreteValues = new[]
                    {
                        0,
                    },
                    MaxValue = 0,
                    MinValue = 0,
                    ScaleType = GoogleNative.Ml.V1.GoogleCloudMlV1__ParameterSpecScaleType.None,
                },
            },
            Algorithm = GoogleNative.Ml.V1.GoogleCloudMlV1__HyperparameterSpecAlgorithm.AlgorithmUnspecified,
            EnableTrialEarlyStopping = false,
            HyperparameterMetricTag = "string",
            MaxFailedTrials = 0,
            MaxParallelTrials = 0,
            MaxTrials = 0,
            ResumePreviousJobId = "string",
        },
        JobDir = "string",
        MasterConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfigArgs
        {
            AcceleratorConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__AcceleratorConfigArgs
            {
                Count = "string",
                Type = GoogleNative.Ml.V1.GoogleCloudMlV1__AcceleratorConfigType.AcceleratorTypeUnspecified,
            },
            ContainerArgs = new[]
            {
                "string",
            },
            ContainerCommand = new[]
            {
                "string",
            },
            DiskConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__DiskConfigArgs
            {
                BootDiskSizeGb = 0,
                BootDiskType = "string",
            },
            ImageUri = "string",
            TpuTfVersion = "string",
        },
        MasterType = "string",
        Network = "string",
        EvaluatorCount = "string",
        Args = new[]
        {
            "string",
        },
        ParameterServerCount = "string",
        ParameterServerType = "string",
        EvaluatorType = "string",
        PythonVersion = "string",
        EncryptionConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__EncryptionConfigArgs
        {
            KmsKeyName = "string",
        },
        RuntimeVersion = "string",
        EnableWebAccess = false,
        Scheduling = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__SchedulingArgs
        {
            MaxRunningTime = "string",
            MaxWaitTime = "string",
            Priority = 0,
        },
        ServiceAccount = "string",
        UseChiefInTfConfig = false,
        WorkerConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfigArgs
        {
            AcceleratorConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__AcceleratorConfigArgs
            {
                Count = "string",
                Type = GoogleNative.Ml.V1.GoogleCloudMlV1__AcceleratorConfigType.AcceleratorTypeUnspecified,
            },
            ContainerArgs = new[]
            {
                "string",
            },
            ContainerCommand = new[]
            {
                "string",
            },
            DiskConfig = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__DiskConfigArgs
            {
                BootDiskSizeGb = 0,
                BootDiskType = "string",
            },
            ImageUri = "string",
            TpuTfVersion = "string",
        },
        WorkerCount = "string",
        WorkerType = "string",
    },
    TrainingOutput = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__TrainingOutputArgs
    {
        BuiltInAlgorithmOutput = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__BuiltInAlgorithmOutputArgs
        {
            Framework = "string",
            ModelPath = "string",
            PythonVersion = "string",
            RuntimeVersion = "string",
        },
        CompletedTrialCount = "string",
        ConsumedMLUnits = 0,
        HyperparameterMetricTag = "string",
        IsBuiltInAlgorithmJob = false,
        IsHyperparameterTuningJob = false,
        Trials = new[]
        {
            new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__HyperparameterOutputArgs
            {
                AllMetrics = new[]
                {
                    new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricArgs
                    {
                        ObjectiveValue = 0,
                        TrainingStep = "string",
                    },
                },
                BuiltInAlgorithmOutput = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__BuiltInAlgorithmOutputArgs
                {
                    Framework = "string",
                    ModelPath = "string",
                    PythonVersion = "string",
                    RuntimeVersion = "string",
                },
                FinalMetric = new GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricArgs
                {
                    ObjectiveValue = 0,
                    TrainingStep = "string",
                },
                Hyperparameters = 
                {
                    { "string", "string" },
                },
                IsTrialStoppedEarly = false,
                TrialId = "string",
                WebAccessUris = 
                {
                    { "string", "string" },
                },
            },
        },
    },
});

    example, err := ml.NewJob(ctx, "examplejobResourceResourceFromMlv1", &ml.JobArgs{
	JobId: pulumi.String("string"),
	Etag:  pulumi.String("string"),
	Labels: pulumi.StringMap{
		"string": pulumi.String("string"),
	},
	PredictionInput: &ml.GoogleCloudMlV1__PredictionInputArgs{
		DataFormat: ml.GoogleCloudMlV1__PredictionInputDataFormatDataFormatUnspecified,
		InputPaths: pulumi.StringArray{
			pulumi.String("string"),
		},
		OutputPath:       pulumi.String("string"),
		Region:           pulumi.String("string"),
		BatchSize:        pulumi.String("string"),
		MaxWorkerCount:   pulumi.String("string"),
		ModelName:        pulumi.String("string"),
		OutputDataFormat: ml.GoogleCloudMlV1__PredictionInputOutputDataFormatDataFormatUnspecified,
		RuntimeVersion:   pulumi.String("string"),
		SignatureName:    pulumi.String("string"),
		Uri:              pulumi.String("string"),
		VersionName:      pulumi.String("string"),
	},
	PredictionOutput: ml.GoogleCloudMlV1__PredictionOutputArgs{
		ErrorCount:      pulumi.String("string"),
		NodeHours:       pulumi.Float64(0),
		OutputPath:      pulumi.String("string"),
		PredictionCount: pulumi.String("string"),
	},
	Project: pulumi.String("string"),
	TrainingInput: &ml.GoogleCloudMlV1__TrainingInputArgs{
		PackageUris: pulumi.StringArray{
			pulumi.String("string"),
		},
		ScaleTier:    ml.GoogleCloudMlV1__TrainingInputScaleTierBasic,
		Region:       pulumi.String("string"),
		PythonModule: pulumi.String("string"),
		ParameterServerConfig: &ml.GoogleCloudMlV1__ReplicaConfigArgs{
			AcceleratorConfig: &ml.GoogleCloudMlV1__AcceleratorConfigArgs{
				Count: pulumi.String("string"),
				Type:  ml.GoogleCloudMlV1__AcceleratorConfigTypeAcceleratorTypeUnspecified,
			},
			ContainerArgs: pulumi.StringArray{
				pulumi.String("string"),
			},
			ContainerCommand: pulumi.StringArray{
				pulumi.String("string"),
			},
			DiskConfig: &ml.GoogleCloudMlV1__DiskConfigArgs{
				BootDiskSizeGb: pulumi.Int(0),
				BootDiskType:   pulumi.String("string"),
			},
			ImageUri:     pulumi.String("string"),
			TpuTfVersion: pulumi.String("string"),
		},
		EvaluatorConfig: &ml.GoogleCloudMlV1__ReplicaConfigArgs{
			AcceleratorConfig: &ml.GoogleCloudMlV1__AcceleratorConfigArgs{
				Count: pulumi.String("string"),
				Type:  ml.GoogleCloudMlV1__AcceleratorConfigTypeAcceleratorTypeUnspecified,
			},
			ContainerArgs: pulumi.StringArray{
				pulumi.String("string"),
			},
			ContainerCommand: pulumi.StringArray{
				pulumi.String("string"),
			},
			DiskConfig: &ml.GoogleCloudMlV1__DiskConfigArgs{
				BootDiskSizeGb: pulumi.Int(0),
				BootDiskType:   pulumi.String("string"),
			},
			ImageUri:     pulumi.String("string"),
			TpuTfVersion: pulumi.String("string"),
		},
		Hyperparameters: &ml.GoogleCloudMlV1__HyperparameterSpecArgs{
			Goal: ml.GoogleCloudMlV1__HyperparameterSpecGoalGoalTypeUnspecified,
			Params: ml.GoogleCloudMlV1__ParameterSpecArray{
				&ml.GoogleCloudMlV1__ParameterSpecArgs{
					ParameterName: pulumi.String("string"),
					Type:          ml.GoogleCloudMlV1__ParameterSpecTypeParameterTypeUnspecified,
					CategoricalValues: pulumi.StringArray{
						pulumi.String("string"),
					},
					DiscreteValues: pulumi.Float64Array{
						pulumi.Float64(0),
					},
					MaxValue:  pulumi.Float64(0),
					MinValue:  pulumi.Float64(0),
					ScaleType: ml.GoogleCloudMlV1__ParameterSpecScaleTypeNone,
				},
			},
			Algorithm:                ml.GoogleCloudMlV1__HyperparameterSpecAlgorithmAlgorithmUnspecified,
			EnableTrialEarlyStopping: pulumi.Bool(false),
			HyperparameterMetricTag:  pulumi.String("string"),
			MaxFailedTrials:          pulumi.Int(0),
			MaxParallelTrials:        pulumi.Int(0),
			MaxTrials:                pulumi.Int(0),
			ResumePreviousJobId:      pulumi.String("string"),
		},
		JobDir: pulumi.String("string"),
		MasterConfig: &ml.GoogleCloudMlV1__ReplicaConfigArgs{
			AcceleratorConfig: &ml.GoogleCloudMlV1__AcceleratorConfigArgs{
				Count: pulumi.String("string"),
				Type:  ml.GoogleCloudMlV1__AcceleratorConfigTypeAcceleratorTypeUnspecified,
			},
			ContainerArgs: pulumi.StringArray{
				pulumi.String("string"),
			},
			ContainerCommand: pulumi.StringArray{
				pulumi.String("string"),
			},
			DiskConfig: &ml.GoogleCloudMlV1__DiskConfigArgs{
				BootDiskSizeGb: pulumi.Int(0),
				BootDiskType:   pulumi.String("string"),
			},
			ImageUri:     pulumi.String("string"),
			TpuTfVersion: pulumi.String("string"),
		},
		MasterType:     pulumi.String("string"),
		Network:        pulumi.String("string"),
		EvaluatorCount: pulumi.String("string"),
		Args: pulumi.StringArray{
			pulumi.String("string"),
		},
		ParameterServerCount: pulumi.String("string"),
		ParameterServerType:  pulumi.String("string"),
		EvaluatorType:        pulumi.String("string"),
		PythonVersion:        pulumi.String("string"),
		EncryptionConfig: &ml.GoogleCloudMlV1__EncryptionConfigArgs{
			KmsKeyName: pulumi.String("string"),
		},
		RuntimeVersion:  pulumi.String("string"),
		EnableWebAccess: pulumi.Bool(false),
		Scheduling: &ml.GoogleCloudMlV1__SchedulingArgs{
			MaxRunningTime: pulumi.String("string"),
			MaxWaitTime:    pulumi.String("string"),
			Priority:       pulumi.Int(0),
		},
		ServiceAccount:     pulumi.String("string"),
		UseChiefInTfConfig: pulumi.Bool(false),
		WorkerConfig: &ml.GoogleCloudMlV1__ReplicaConfigArgs{
			AcceleratorConfig: &ml.GoogleCloudMlV1__AcceleratorConfigArgs{
				Count: pulumi.String("string"),
				Type:  ml.GoogleCloudMlV1__AcceleratorConfigTypeAcceleratorTypeUnspecified,
			},
			ContainerArgs: pulumi.StringArray{
				pulumi.String("string"),
			},
			ContainerCommand: pulumi.StringArray{
				pulumi.String("string"),
			},
			DiskConfig: &ml.GoogleCloudMlV1__DiskConfigArgs{
				BootDiskSizeGb: pulumi.Int(0),
				BootDiskType:   pulumi.String("string"),
			},
			ImageUri:     pulumi.String("string"),
			TpuTfVersion: pulumi.String("string"),
		},
		WorkerCount: pulumi.String("string"),
		WorkerType:  pulumi.String("string"),
	},
	TrainingOutput: ml.GoogleCloudMlV1__TrainingOutputArgs{
		BuiltInAlgorithmOutput: ml.GoogleCloudMlV1__BuiltInAlgorithmOutputArgs{
			Framework:      pulumi.String("string"),
			ModelPath:      pulumi.String("string"),
			PythonVersion:  pulumi.String("string"),
			RuntimeVersion: pulumi.String("string"),
		},
		CompletedTrialCount:       pulumi.String("string"),
		ConsumedMLUnits:           pulumi.Float64(0),
		HyperparameterMetricTag:   pulumi.String("string"),
		IsBuiltInAlgorithmJob:     pulumi.Bool(false),
		IsHyperparameterTuningJob: pulumi.Bool(false),
		Trials: ml.GoogleCloudMlV1__HyperparameterOutputArray{
			ml.GoogleCloudMlV1__HyperparameterOutputArgs{
				AllMetrics: ml.GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricArray{
					&ml.GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricArgs{
						ObjectiveValue: pulumi.Float64(0),
						TrainingStep:   pulumi.String("string"),
					},
				},
				BuiltInAlgorithmOutput: ml.GoogleCloudMlV1__BuiltInAlgorithmOutputArgs{
					Framework:      pulumi.String("string"),
					ModelPath:      pulumi.String("string"),
					PythonVersion:  pulumi.String("string"),
					RuntimeVersion: pulumi.String("string"),
				},
				FinalMetric: &ml.GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricArgs{
					ObjectiveValue: pulumi.Float64(0),
					TrainingStep:   pulumi.String("string"),
				},
				Hyperparameters: pulumi.StringMap{
					"string": pulumi.String("string"),
				},
				IsTrialStoppedEarly: pulumi.Bool(false),
				TrialId:             pulumi.String("string"),
				WebAccessUris: pulumi.StringMap{
					"string": pulumi.String("string"),
				},
			},
		},
	},
})

    var examplejobResourceResourceFromMlv1 = new Job("examplejobResourceResourceFromMlv1", JobArgs.builder()
    .jobId("string")
    .etag("string")
    .labels(Map.of("string", "string"))
    .predictionInput(GoogleCloudMlV1__PredictionInputArgs.builder()
        .dataFormat("DATA_FORMAT_UNSPECIFIED")
        .inputPaths("string")
        .outputPath("string")
        .region("string")
        .batchSize("string")
        .maxWorkerCount("string")
        .modelName("string")
        .outputDataFormat("DATA_FORMAT_UNSPECIFIED")
        .runtimeVersion("string")
        .signatureName("string")
        .uri("string")
        .versionName("string")
        .build())
    .predictionOutput(GoogleCloudMlV1__PredictionOutputArgs.builder()
        .errorCount("string")
        .nodeHours(0)
        .outputPath("string")
        .predictionCount("string")
        .build())
    .project("string")
    .trainingInput(GoogleCloudMlV1__TrainingInputArgs.builder()
        .packageUris("string")
        .scaleTier("BASIC")
        .region("string")
        .pythonModule("string")
        .parameterServerConfig(GoogleCloudMlV1__ReplicaConfigArgs.builder()
            .acceleratorConfig(GoogleCloudMlV1__AcceleratorConfigArgs.builder()
                .count("string")
                .type("ACCELERATOR_TYPE_UNSPECIFIED")
                .build())
            .containerArgs("string")
            .containerCommand("string")
            .diskConfig(GoogleCloudMlV1__DiskConfigArgs.builder()
                .bootDiskSizeGb(0)
                .bootDiskType("string")
                .build())
            .imageUri("string")
            .tpuTfVersion("string")
            .build())
        .evaluatorConfig(GoogleCloudMlV1__ReplicaConfigArgs.builder()
            .acceleratorConfig(GoogleCloudMlV1__AcceleratorConfigArgs.builder()
                .count("string")
                .type("ACCELERATOR_TYPE_UNSPECIFIED")
                .build())
            .containerArgs("string")
            .containerCommand("string")
            .diskConfig(GoogleCloudMlV1__DiskConfigArgs.builder()
                .bootDiskSizeGb(0)
                .bootDiskType("string")
                .build())
            .imageUri("string")
            .tpuTfVersion("string")
            .build())
        .hyperparameters(GoogleCloudMlV1__HyperparameterSpecArgs.builder()
            .goal("GOAL_TYPE_UNSPECIFIED")
            .params(GoogleCloudMlV1__ParameterSpecArgs.builder()
                .parameterName("string")
                .type("PARAMETER_TYPE_UNSPECIFIED")
                .categoricalValues("string")
                .discreteValues(0)
                .maxValue(0)
                .minValue(0)
                .scaleType("NONE")
                .build())
            .algorithm("ALGORITHM_UNSPECIFIED")
            .enableTrialEarlyStopping(false)
            .hyperparameterMetricTag("string")
            .maxFailedTrials(0)
            .maxParallelTrials(0)
            .maxTrials(0)
            .resumePreviousJobId("string")
            .build())
        .jobDir("string")
        .masterConfig(GoogleCloudMlV1__ReplicaConfigArgs.builder()
            .acceleratorConfig(GoogleCloudMlV1__AcceleratorConfigArgs.builder()
                .count("string")
                .type("ACCELERATOR_TYPE_UNSPECIFIED")
                .build())
            .containerArgs("string")
            .containerCommand("string")
            .diskConfig(GoogleCloudMlV1__DiskConfigArgs.builder()
                .bootDiskSizeGb(0)
                .bootDiskType("string")
                .build())
            .imageUri("string")
            .tpuTfVersion("string")
            .build())
        .masterType("string")
        .network("string")
        .evaluatorCount("string")
        .args("string")
        .parameterServerCount("string")
        .parameterServerType("string")
        .evaluatorType("string")
        .pythonVersion("string")
        .encryptionConfig(GoogleCloudMlV1__EncryptionConfigArgs.builder()
            .kmsKeyName("string")
            .build())
        .runtimeVersion("string")
        .enableWebAccess(false)
        .scheduling(GoogleCloudMlV1__SchedulingArgs.builder()
            .maxRunningTime("string")
            .maxWaitTime("string")
            .priority(0)
            .build())
        .serviceAccount("string")
        .useChiefInTfConfig(false)
        .workerConfig(GoogleCloudMlV1__ReplicaConfigArgs.builder()
            .acceleratorConfig(GoogleCloudMlV1__AcceleratorConfigArgs.builder()
                .count("string")
                .type("ACCELERATOR_TYPE_UNSPECIFIED")
                .build())
            .containerArgs("string")
            .containerCommand("string")
            .diskConfig(GoogleCloudMlV1__DiskConfigArgs.builder()
                .bootDiskSizeGb(0)
                .bootDiskType("string")
                .build())
            .imageUri("string")
            .tpuTfVersion("string")
            .build())
        .workerCount("string")
        .workerType("string")
        .build())
    .trainingOutput(GoogleCloudMlV1__TrainingOutputArgs.builder()
        .builtInAlgorithmOutput(GoogleCloudMlV1__BuiltInAlgorithmOutputArgs.builder()
            .framework("string")
            .modelPath("string")
            .pythonVersion("string")
            .runtimeVersion("string")
            .build())
        .completedTrialCount("string")
        .consumedMLUnits(0)
        .hyperparameterMetricTag("string")
        .isBuiltInAlgorithmJob(false)
        .isHyperparameterTuningJob(false)
        .trials(GoogleCloudMlV1__HyperparameterOutputArgs.builder()
            .allMetrics(GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricArgs.builder()
                .objectiveValue(0)
                .trainingStep("string")
                .build())
            .builtInAlgorithmOutput(GoogleCloudMlV1__BuiltInAlgorithmOutputArgs.builder()
                .framework("string")
                .modelPath("string")
                .pythonVersion("string")
                .runtimeVersion("string")
                .build())
            .finalMetric(GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricArgs.builder()
                .objectiveValue(0)
                .trainingStep("string")
                .build())
            .hyperparameters(Map.of("string", "string"))
            .isTrialStoppedEarly(false)
            .trialId("string")
            .webAccessUris(Map.of("string", "string"))
            .build())
        .build())
    .build());

    examplejob_resource_resource_from_mlv1 = google_native.ml.v1.Job("examplejobResourceResourceFromMlv1",
    job_id="string",
    etag="string",
    labels={
        "string": "string",
    },
    prediction_input={
        "data_format": google_native.ml.v1.GoogleCloudMlV1__PredictionInputDataFormat.DATA_FORMAT_UNSPECIFIED,
        "input_paths": ["string"],
        "output_path": "string",
        "region": "string",
        "batch_size": "string",
        "max_worker_count": "string",
        "model_name": "string",
        "output_data_format": google_native.ml.v1.GoogleCloudMlV1__PredictionInputOutputDataFormat.DATA_FORMAT_UNSPECIFIED,
        "runtime_version": "string",
        "signature_name": "string",
        "uri": "string",
        "version_name": "string",
    },
    prediction_output={
        "error_count": "string",
        "node_hours": 0,
        "output_path": "string",
        "prediction_count": "string",
    },
    project="string",
    training_input={
        "package_uris": ["string"],
        "scale_tier": google_native.ml.v1.GoogleCloudMlV1__TrainingInputScaleTier.BASIC,
        "region": "string",
        "python_module": "string",
        "parameter_server_config": {
            "accelerator_config": {
                "count": "string",
                "type": google_native.ml.v1.GoogleCloudMlV1__AcceleratorConfigType.ACCELERATOR_TYPE_UNSPECIFIED,
            },
            "container_args": ["string"],
            "container_command": ["string"],
            "disk_config": {
                "boot_disk_size_gb": 0,
                "boot_disk_type": "string",
            },
            "image_uri": "string",
            "tpu_tf_version": "string",
        },
        "evaluator_config": {
            "accelerator_config": {
                "count": "string",
                "type": google_native.ml.v1.GoogleCloudMlV1__AcceleratorConfigType.ACCELERATOR_TYPE_UNSPECIFIED,
            },
            "container_args": ["string"],
            "container_command": ["string"],
            "disk_config": {
                "boot_disk_size_gb": 0,
                "boot_disk_type": "string",
            },
            "image_uri": "string",
            "tpu_tf_version": "string",
        },
        "hyperparameters": {
            "goal": google_native.ml.v1.GoogleCloudMlV1__HyperparameterSpecGoal.GOAL_TYPE_UNSPECIFIED,
            "params": [{
                "parameter_name": "string",
                "type": google_native.ml.v1.GoogleCloudMlV1__ParameterSpecType.PARAMETER_TYPE_UNSPECIFIED,
                "categorical_values": ["string"],
                "discrete_values": [0],
                "max_value": 0,
                "min_value": 0,
                "scale_type": google_native.ml.v1.GoogleCloudMlV1__ParameterSpecScaleType.NONE,
            }],
            "algorithm": google_native.ml.v1.GoogleCloudMlV1__HyperparameterSpecAlgorithm.ALGORITHM_UNSPECIFIED,
            "enable_trial_early_stopping": False,
            "hyperparameter_metric_tag": "string",
            "max_failed_trials": 0,
            "max_parallel_trials": 0,
            "max_trials": 0,
            "resume_previous_job_id": "string",
        },
        "job_dir": "string",
        "master_config": {
            "accelerator_config": {
                "count": "string",
                "type": google_native.ml.v1.GoogleCloudMlV1__AcceleratorConfigType.ACCELERATOR_TYPE_UNSPECIFIED,
            },
            "container_args": ["string"],
            "container_command": ["string"],
            "disk_config": {
                "boot_disk_size_gb": 0,
                "boot_disk_type": "string",
            },
            "image_uri": "string",
            "tpu_tf_version": "string",
        },
        "master_type": "string",
        "network": "string",
        "evaluator_count": "string",
        "args": ["string"],
        "parameter_server_count": "string",
        "parameter_server_type": "string",
        "evaluator_type": "string",
        "python_version": "string",
        "encryption_config": {
            "kms_key_name": "string",
        },
        "runtime_version": "string",
        "enable_web_access": False,
        "scheduling": {
            "max_running_time": "string",
            "max_wait_time": "string",
            "priority": 0,
        },
        "service_account": "string",
        "use_chief_in_tf_config": False,
        "worker_config": {
            "accelerator_config": {
                "count": "string",
                "type": google_native.ml.v1.GoogleCloudMlV1__AcceleratorConfigType.ACCELERATOR_TYPE_UNSPECIFIED,
            },
            "container_args": ["string"],
            "container_command": ["string"],
            "disk_config": {
                "boot_disk_size_gb": 0,
                "boot_disk_type": "string",
            },
            "image_uri": "string",
            "tpu_tf_version": "string",
        },
        "worker_count": "string",
        "worker_type": "string",
    },
    training_output={
        "built_in_algorithm_output": {
            "framework": "string",
            "model_path": "string",
            "python_version": "string",
            "runtime_version": "string",
        },
        "completed_trial_count": "string",
        "consumed_ml_units": 0,
        "hyperparameter_metric_tag": "string",
        "is_built_in_algorithm_job": False,
        "is_hyperparameter_tuning_job": False,
        "trials": [{
            "all_metrics": [{
                "objective_value": 0,
                "training_step": "string",
            }],
            "built_in_algorithm_output": {
                "framework": "string",
                "model_path": "string",
                "python_version": "string",
                "runtime_version": "string",
            },
            "final_metric": {
                "objective_value": 0,
                "training_step": "string",
            },
            "hyperparameters": {
                "string": "string",
            },
            "is_trial_stopped_early": False,
            "trial_id": "string",
            "web_access_uris": {
                "string": "string",
            },
        }],
    })

    const examplejobResourceResourceFromMlv1 = new google_native.ml.v1.Job("examplejobResourceResourceFromMlv1", {
    jobId: "string",
    etag: "string",
    labels: {
        string: "string",
    },
    predictionInput: {
        dataFormat: google_native.ml.v1.GoogleCloudMlV1__PredictionInputDataFormat.DataFormatUnspecified,
        inputPaths: ["string"],
        outputPath: "string",
        region: "string",
        batchSize: "string",
        maxWorkerCount: "string",
        modelName: "string",
        outputDataFormat: google_native.ml.v1.GoogleCloudMlV1__PredictionInputOutputDataFormat.DataFormatUnspecified,
        runtimeVersion: "string",
        signatureName: "string",
        uri: "string",
        versionName: "string",
    },
    predictionOutput: {
        errorCount: "string",
        nodeHours: 0,
        outputPath: "string",
        predictionCount: "string",
    },
    project: "string",
    trainingInput: {
        packageUris: ["string"],
        scaleTier: google_native.ml.v1.GoogleCloudMlV1__TrainingInputScaleTier.Basic,
        region: "string",
        pythonModule: "string",
        parameterServerConfig: {
            acceleratorConfig: {
                count: "string",
                type: google_native.ml.v1.GoogleCloudMlV1__AcceleratorConfigType.AcceleratorTypeUnspecified,
            },
            containerArgs: ["string"],
            containerCommand: ["string"],
            diskConfig: {
                bootDiskSizeGb: 0,
                bootDiskType: "string",
            },
            imageUri: "string",
            tpuTfVersion: "string",
        },
        evaluatorConfig: {
            acceleratorConfig: {
                count: "string",
                type: google_native.ml.v1.GoogleCloudMlV1__AcceleratorConfigType.AcceleratorTypeUnspecified,
            },
            containerArgs: ["string"],
            containerCommand: ["string"],
            diskConfig: {
                bootDiskSizeGb: 0,
                bootDiskType: "string",
            },
            imageUri: "string",
            tpuTfVersion: "string",
        },
        hyperparameters: {
            goal: google_native.ml.v1.GoogleCloudMlV1__HyperparameterSpecGoal.GoalTypeUnspecified,
            params: [{
                parameterName: "string",
                type: google_native.ml.v1.GoogleCloudMlV1__ParameterSpecType.ParameterTypeUnspecified,
                categoricalValues: ["string"],
                discreteValues: [0],
                maxValue: 0,
                minValue: 0,
                scaleType: google_native.ml.v1.GoogleCloudMlV1__ParameterSpecScaleType.None,
            }],
            algorithm: google_native.ml.v1.GoogleCloudMlV1__HyperparameterSpecAlgorithm.AlgorithmUnspecified,
            enableTrialEarlyStopping: false,
            hyperparameterMetricTag: "string",
            maxFailedTrials: 0,
            maxParallelTrials: 0,
            maxTrials: 0,
            resumePreviousJobId: "string",
        },
        jobDir: "string",
        masterConfig: {
            acceleratorConfig: {
                count: "string",
                type: google_native.ml.v1.GoogleCloudMlV1__AcceleratorConfigType.AcceleratorTypeUnspecified,
            },
            containerArgs: ["string"],
            containerCommand: ["string"],
            diskConfig: {
                bootDiskSizeGb: 0,
                bootDiskType: "string",
            },
            imageUri: "string",
            tpuTfVersion: "string",
        },
        masterType: "string",
        network: "string",
        evaluatorCount: "string",
        args: ["string"],
        parameterServerCount: "string",
        parameterServerType: "string",
        evaluatorType: "string",
        pythonVersion: "string",
        encryptionConfig: {
            kmsKeyName: "string",
        },
        runtimeVersion: "string",
        enableWebAccess: false,
        scheduling: {
            maxRunningTime: "string",
            maxWaitTime: "string",
            priority: 0,
        },
        serviceAccount: "string",
        useChiefInTfConfig: false,
        workerConfig: {
            acceleratorConfig: {
                count: "string",
                type: google_native.ml.v1.GoogleCloudMlV1__AcceleratorConfigType.AcceleratorTypeUnspecified,
            },
            containerArgs: ["string"],
            containerCommand: ["string"],
            diskConfig: {
                bootDiskSizeGb: 0,
                bootDiskType: "string",
            },
            imageUri: "string",
            tpuTfVersion: "string",
        },
        workerCount: "string",
        workerType: "string",
    },
    trainingOutput: {
        builtInAlgorithmOutput: {
            framework: "string",
            modelPath: "string",
            pythonVersion: "string",
            runtimeVersion: "string",
        },
        completedTrialCount: "string",
        consumedMLUnits: 0,
        hyperparameterMetricTag: "string",
        isBuiltInAlgorithmJob: false,
        isHyperparameterTuningJob: false,
        trials: [{
            allMetrics: [{
                objectiveValue: 0,
                trainingStep: "string",
            }],
            builtInAlgorithmOutput: {
                framework: "string",
                modelPath: "string",
                pythonVersion: "string",
                runtimeVersion: "string",
            },
            finalMetric: {
                objectiveValue: 0,
                trainingStep: "string",
            },
            hyperparameters: {
                string: "string",
            },
            isTrialStoppedEarly: false,
            trialId: "string",
            webAccessUris: {
                string: "string",
            },
        }],
    },
});

    type: google-native:ml/v1:Job
properties:
    etag: string
    jobId: string
    labels:
        string: string
    predictionInput:
        batchSize: string
        dataFormat: DATA_FORMAT_UNSPECIFIED
        inputPaths:
            - string
        maxWorkerCount: string
        modelName: string
        outputDataFormat: DATA_FORMAT_UNSPECIFIED
        outputPath: string
        region: string
        runtimeVersion: string
        signatureName: string
        uri: string
        versionName: string
    predictionOutput:
        errorCount: string
        nodeHours: 0
        outputPath: string
        predictionCount: string
    project: string
    trainingInput:
        args:
            - string
        enableWebAccess: false
        encryptionConfig:
            kmsKeyName: string
        evaluatorConfig:
            acceleratorConfig:
                count: string
                type: ACCELERATOR_TYPE_UNSPECIFIED
            containerArgs:
                - string
            containerCommand:
                - string
            diskConfig:
                bootDiskSizeGb: 0
                bootDiskType: string
            imageUri: string
            tpuTfVersion: string
        evaluatorCount: string
        evaluatorType: string
        hyperparameters:
            algorithm: ALGORITHM_UNSPECIFIED
            enableTrialEarlyStopping: false
            goal: GOAL_TYPE_UNSPECIFIED
            hyperparameterMetricTag: string
            maxFailedTrials: 0
            maxParallelTrials: 0
            maxTrials: 0
            params:
                - categoricalValues:
                    - string
                  discreteValues:
                    - 0
                  maxValue: 0
                  minValue: 0
                  parameterName: string
                  scaleType: NONE
                  type: PARAMETER_TYPE_UNSPECIFIED
            resumePreviousJobId: string
        jobDir: string
        masterConfig:
            acceleratorConfig:
                count: string
                type: ACCELERATOR_TYPE_UNSPECIFIED
            containerArgs:
                - string
            containerCommand:
                - string
            diskConfig:
                bootDiskSizeGb: 0
                bootDiskType: string
            imageUri: string
            tpuTfVersion: string
        masterType: string
        network: string
        packageUris:
            - string
        parameterServerConfig:
            acceleratorConfig:
                count: string
                type: ACCELERATOR_TYPE_UNSPECIFIED
            containerArgs:
                - string
            containerCommand:
                - string
            diskConfig:
                bootDiskSizeGb: 0
                bootDiskType: string
            imageUri: string
            tpuTfVersion: string
        parameterServerCount: string
        parameterServerType: string
        pythonModule: string
        pythonVersion: string
        region: string
        runtimeVersion: string
        scaleTier: BASIC
        scheduling:
            maxRunningTime: string
            maxWaitTime: string
            priority: 0
        serviceAccount: string
        useChiefInTfConfig: false
        workerConfig:
            acceleratorConfig:
                count: string
                type: ACCELERATOR_TYPE_UNSPECIFIED
            containerArgs:
                - string
            containerCommand:
                - string
            diskConfig:
                bootDiskSizeGb: 0
                bootDiskType: string
            imageUri: string
            tpuTfVersion: string
        workerCount: string
        workerType: string
    trainingOutput:
        builtInAlgorithmOutput:
            framework: string
            modelPath: string
            pythonVersion: string
            runtimeVersion: string
        completedTrialCount: string
        consumedMLUnits: 0
        hyperparameterMetricTag: string
        isBuiltInAlgorithmJob: false
        isHyperparameterTuningJob: false
        trials:
            - allMetrics:
                - objectiveValue: 0
                  trainingStep: string
              builtInAlgorithmOutput:
                framework: string
                modelPath: string
                pythonVersion: string
                runtimeVersion: string
              finalMetric:
                objectiveValue: 0
                trainingStep: string
              hyperparameters:
                string: string
              isTrialStoppedEarly: false
              trialId: string
              webAccessUris:
                string: string

    Job Resource Properties

    To learn more about resource properties and how to use them, see Inputs and Outputs in the Architecture and Concepts docs.

    Inputs

    In Python, inputs that are objects can be passed either as argument classes or as dictionary literals.

    The Job resource accepts the following input properties:

    JobId string
    The user-specified id of the job.
    Etag string
    etag is used for optimistic concurrency control as a way to help prevent simultaneous updates of a job from overwriting each other. It is strongly suggested that systems make use of the etag in the read-modify-write cycle to perform job updates in order to avoid race conditions: An etag is returned in the response to GetJob, and systems are expected to put that etag in the request to UpdateJob to ensure that their change will be applied to the same version of the job.
    Labels Dictionary<string, string>
    Optional. One or more labels that you can add, to organize your jobs. Each label is a key-value pair, where both the key and the value are arbitrary strings that you supply. For more information, see the documentation on using labels.
    PredictionInput Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__PredictionInput
    Input parameters to create a prediction job.
    PredictionOutput Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__PredictionOutput
    The current prediction job result.
    Project string
    TrainingInput Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__TrainingInput
    Input parameters to create a training job.
    TrainingOutput Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__TrainingOutput
    The current training job result.
    JobId string
    The user-specified id of the job.
    Etag string
    etag is used for optimistic concurrency control as a way to help prevent simultaneous updates of a job from overwriting each other. It is strongly suggested that systems make use of the etag in the read-modify-write cycle to perform job updates in order to avoid race conditions: An etag is returned in the response to GetJob, and systems are expected to put that etag in the request to UpdateJob to ensure that their change will be applied to the same version of the job.
    Labels map[string]string
    Optional. One or more labels that you can add, to organize your jobs. Each label is a key-value pair, where both the key and the value are arbitrary strings that you supply. For more information, see the documentation on using labels.
    PredictionInput GoogleCloudMlV1__PredictionInputArgs
    Input parameters to create a prediction job.
    PredictionOutput GoogleCloudMlV1__PredictionOutputArgs
    The current prediction job result.
    Project string
    TrainingInput GoogleCloudMlV1__TrainingInputArgs
    Input parameters to create a training job.
    TrainingOutput GoogleCloudMlV1__TrainingOutputArgs
    The current training job result.
    jobId String
    The user-specified id of the job.
    etag String
    etag is used for optimistic concurrency control as a way to help prevent simultaneous updates of a job from overwriting each other. It is strongly suggested that systems make use of the etag in the read-modify-write cycle to perform job updates in order to avoid race conditions: An etag is returned in the response to GetJob, and systems are expected to put that etag in the request to UpdateJob to ensure that their change will be applied to the same version of the job.
    labels Map<String,String>
    Optional. One or more labels that you can add, to organize your jobs. Each label is a key-value pair, where both the key and the value are arbitrary strings that you supply. For more information, see the documentation on using labels.
    predictionInput GoogleCloudMlV1__PredictionInput
    Input parameters to create a prediction job.
    predictionOutput GoogleCloudMlV1__PredictionOutput
    The current prediction job result.
    project String
    trainingInput GoogleCloudMlV1__TrainingInput
    Input parameters to create a training job.
    trainingOutput GoogleCloudMlV1__TrainingOutput
    The current training job result.
    jobId string
    The user-specified id of the job.
    etag string
    etag is used for optimistic concurrency control as a way to help prevent simultaneous updates of a job from overwriting each other. It is strongly suggested that systems make use of the etag in the read-modify-write cycle to perform job updates in order to avoid race conditions: An etag is returned in the response to GetJob, and systems are expected to put that etag in the request to UpdateJob to ensure that their change will be applied to the same version of the job.
    labels {[key: string]: string}
    Optional. One or more labels that you can add, to organize your jobs. Each label is a key-value pair, where both the key and the value are arbitrary strings that you supply. For more information, see the documentation on using labels.
    predictionInput GoogleCloudMlV1__PredictionInput
    Input parameters to create a prediction job.
    predictionOutput GoogleCloudMlV1__PredictionOutput
    The current prediction job result.
    project string
    trainingInput GoogleCloudMlV1__TrainingInput
    Input parameters to create a training job.
    trainingOutput GoogleCloudMlV1__TrainingOutput
    The current training job result.
    job_id str
    The user-specified id of the job.
    etag str
    etag is used for optimistic concurrency control as a way to help prevent simultaneous updates of a job from overwriting each other. It is strongly suggested that systems make use of the etag in the read-modify-write cycle to perform job updates in order to avoid race conditions: An etag is returned in the response to GetJob, and systems are expected to put that etag in the request to UpdateJob to ensure that their change will be applied to the same version of the job.
    labels Mapping[str, str]
    Optional. One or more labels that you can add, to organize your jobs. Each label is a key-value pair, where both the key and the value are arbitrary strings that you supply. For more information, see the documentation on using labels.
    prediction_input GoogleCloudMlV1PredictionInputArgs
    Input parameters to create a prediction job.
    prediction_output GoogleCloudMlV1PredictionOutputArgs
    The current prediction job result.
    project str
    training_input GoogleCloudMlV1TrainingInputArgs
    Input parameters to create a training job.
    training_output GoogleCloudMlV1TrainingOutputArgs
    The current training job result.
    jobId String
    The user-specified id of the job.
    etag String
    etag is used for optimistic concurrency control as a way to help prevent simultaneous updates of a job from overwriting each other. It is strongly suggested that systems make use of the etag in the read-modify-write cycle to perform job updates in order to avoid race conditions: An etag is returned in the response to GetJob, and systems are expected to put that etag in the request to UpdateJob to ensure that their change will be applied to the same version of the job.
    labels Map<String>
    Optional. One or more labels that you can add, to organize your jobs. Each label is a key-value pair, where both the key and the value are arbitrary strings that you supply. For more information, see the documentation on using labels.
    predictionInput Property Map
    Input parameters to create a prediction job.
    predictionOutput Property Map
    The current prediction job result.
    project String
    trainingInput Property Map
    Input parameters to create a training job.
    trainingOutput Property Map
    The current training job result.

    Outputs

    All input properties are implicitly available as output properties. Additionally, the Job resource produces the following output properties:

    CreateTime string
    When the job was created.
    EndTime string
    When the job processing was completed.
    ErrorMessage string
    The details of a failure or a cancellation.
    Id string
    The provider-assigned unique ID for this managed resource.
    JobPosition string
    It's only effect when the job is in QUEUED state. If it's positive, it indicates the job's position in the job scheduler. It's 0 when the job is already scheduled.
    StartTime string
    When the job processing was started.
    State string
    The detailed state of a job.
    CreateTime string
    When the job was created.
    EndTime string
    When the job processing was completed.
    ErrorMessage string
    The details of a failure or a cancellation.
    Id string
    The provider-assigned unique ID for this managed resource.
    JobPosition string
    It's only effect when the job is in QUEUED state. If it's positive, it indicates the job's position in the job scheduler. It's 0 when the job is already scheduled.
    StartTime string
    When the job processing was started.
    State string
    The detailed state of a job.
    createTime String
    When the job was created.
    endTime String
    When the job processing was completed.
    errorMessage String
    The details of a failure or a cancellation.
    id String
    The provider-assigned unique ID for this managed resource.
    jobPosition String
    It's only effect when the job is in QUEUED state. If it's positive, it indicates the job's position in the job scheduler. It's 0 when the job is already scheduled.
    startTime String
    When the job processing was started.
    state String
    The detailed state of a job.
    createTime string
    When the job was created.
    endTime string
    When the job processing was completed.
    errorMessage string
    The details of a failure or a cancellation.
    id string
    The provider-assigned unique ID for this managed resource.
    jobPosition string
    It's only effect when the job is in QUEUED state. If it's positive, it indicates the job's position in the job scheduler. It's 0 when the job is already scheduled.
    startTime string
    When the job processing was started.
    state string
    The detailed state of a job.
    create_time str
    When the job was created.
    end_time str
    When the job processing was completed.
    error_message str
    The details of a failure or a cancellation.
    id str
    The provider-assigned unique ID for this managed resource.
    job_position str
    It's only effect when the job is in QUEUED state. If it's positive, it indicates the job's position in the job scheduler. It's 0 when the job is already scheduled.
    start_time str
    When the job processing was started.
    state str
    The detailed state of a job.
    createTime String
    When the job was created.
    endTime String
    When the job processing was completed.
    errorMessage String
    The details of a failure or a cancellation.
    id String
    The provider-assigned unique ID for this managed resource.
    jobPosition String
    It's only effect when the job is in QUEUED state. If it's positive, it indicates the job's position in the job scheduler. It's 0 when the job is already scheduled.
    startTime String
    When the job processing was started.
    state String
    The detailed state of a job.

    Supporting Types

    GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetric, GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricArgs

    ObjectiveValue double
    The objective value at this training step.
    TrainingStep string
    The global training step for this metric.
    ObjectiveValue float64
    The objective value at this training step.
    TrainingStep string
    The global training step for this metric.
    objectiveValue Double
    The objective value at this training step.
    trainingStep String
    The global training step for this metric.
    objectiveValue number
    The objective value at this training step.
    trainingStep string
    The global training step for this metric.
    objective_value float
    The objective value at this training step.
    training_step str
    The global training step for this metric.
    objectiveValue Number
    The objective value at this training step.
    trainingStep String
    The global training step for this metric.

    GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponse, GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponseArgs

    ObjectiveValue double
    The objective value at this training step.
    TrainingStep string
    The global training step for this metric.
    ObjectiveValue float64
    The objective value at this training step.
    TrainingStep string
    The global training step for this metric.
    objectiveValue Double
    The objective value at this training step.
    trainingStep String
    The global training step for this metric.
    objectiveValue number
    The objective value at this training step.
    trainingStep string
    The global training step for this metric.
    objective_value float
    The objective value at this training step.
    training_step str
    The global training step for this metric.
    objectiveValue Number
    The objective value at this training step.
    trainingStep String
    The global training step for this metric.

    GoogleCloudMlV1__AcceleratorConfig, GoogleCloudMlV1__AcceleratorConfigArgs

    Count string
    The number of accelerators to attach to each machine running the job.
    Type Pulumi.GoogleNative.Ml.V1.GoogleCloudMlV1__AcceleratorConfigType
    The type of accelerator to use.
    Count string
    The number of accelerators to attach to each machine running the job.
    Type GoogleCloudMlV1__AcceleratorConfigType
    The type of accelerator to use.
    count String
    The number of accelerators to attach to each machine running the job.
    type GoogleCloudMlV1__AcceleratorConfigType
    The type of accelerator to use.
    count string
    The number of accelerators to attach to each machine running the job.
    type GoogleCloudMlV1__AcceleratorConfigType
    The type of accelerator to use.
    count str
    The number of accelerators to attach to each machine running the job.
    type GoogleCloudMlV1AcceleratorConfigType
    The type of accelerator to use.

    GoogleCloudMlV1__AcceleratorConfigResponse, GoogleCloudMlV1__AcceleratorConfigResponseArgs

    Count string
    The number of accelerators to attach to each machine running the job.
    Type string
    The type of accelerator to use.
    Count string
    The number of accelerators to attach to each machine running the job.
    Type string
    The type of accelerator to use.
    count String
    The number of accelerators to attach to each machine running the job.
    type String
    The type of accelerator to use.
    count string
    The number of accelerators to attach to each machine running the job.
    type string
    The type of accelerator to use.
    count str
    The number of accelerators to attach to each machine running the job.
    type str
    The type of accelerator to use.
    count String
    The number of accelerators to attach to each machine running the job.
    type String
    The type of accelerator to use.

    GoogleCloudMlV1__AcceleratorConfigType, GoogleCloudMlV1__AcceleratorConfigTypeArgs

    AcceleratorTypeUnspecified
    ACCELERATOR_TYPE_UNSPECIFIEDUnspecified accelerator type. Default to no GPU.
    NvidiaTeslaK80
    NVIDIA_TESLA_K80Nvidia Tesla K80 GPU.
    NvidiaTeslaP100
    NVIDIA_TESLA_P100Nvidia Tesla P100 GPU.
    NvidiaTeslaV100
    NVIDIA_TESLA_V100Nvidia V100 GPU.
    NvidiaTeslaP4
    NVIDIA_TESLA_P4Nvidia Tesla P4 GPU.
    NvidiaTeslaT4
    NVIDIA_TESLA_T4Nvidia T4 GPU.
    NvidiaTeslaA100
    NVIDIA_TESLA_A100Nvidia A100 GPU.
    TpuV2
    TPU_V2TPU v2.
    TpuV3
    TPU_V3TPU v3.
    TpuV2Pod
    TPU_V2_PODTPU v2 POD.
    TpuV3Pod
    TPU_V3_PODTPU v3 POD.
    TpuV4Pod
    TPU_V4_PODTPU v4 POD.
    GoogleCloudMlV1__AcceleratorConfigTypeAcceleratorTypeUnspecified
    ACCELERATOR_TYPE_UNSPECIFIEDUnspecified accelerator type. Default to no GPU.
    GoogleCloudMlV1__AcceleratorConfigTypeNvidiaTeslaK80
    NVIDIA_TESLA_K80Nvidia Tesla K80 GPU.
    GoogleCloudMlV1__AcceleratorConfigTypeNvidiaTeslaP100
    NVIDIA_TESLA_P100Nvidia Tesla P100 GPU.
    GoogleCloudMlV1__AcceleratorConfigTypeNvidiaTeslaV100
    NVIDIA_TESLA_V100Nvidia V100 GPU.
    GoogleCloudMlV1__AcceleratorConfigTypeNvidiaTeslaP4
    NVIDIA_TESLA_P4Nvidia Tesla P4 GPU.
    GoogleCloudMlV1__AcceleratorConfigTypeNvidiaTeslaT4
    NVIDIA_TESLA_T4Nvidia T4 GPU.
    GoogleCloudMlV1__AcceleratorConfigTypeNvidiaTeslaA100
    NVIDIA_TESLA_A100Nvidia A100 GPU.
    GoogleCloudMlV1__AcceleratorConfigTypeTpuV2
    TPU_V2TPU v2.
    GoogleCloudMlV1__AcceleratorConfigTypeTpuV3
    TPU_V3TPU v3.
    GoogleCloudMlV1__AcceleratorConfigTypeTpuV2Pod
    TPU_V2_PODTPU v2 POD.
    GoogleCloudMlV1__AcceleratorConfigTypeTpuV3Pod
    TPU_V3_PODTPU v3 POD.
    GoogleCloudMlV1__AcceleratorConfigTypeTpuV4Pod
    TPU_V4_PODTPU v4 POD.
    AcceleratorTypeUnspecified
    ACCELERATOR_TYPE_UNSPECIFIEDUnspecified accelerator type. Default to no GPU.
    NvidiaTeslaK80
    NVIDIA_TESLA_K80Nvidia Tesla K80 GPU.
    NvidiaTeslaP100
    NVIDIA_TESLA_P100Nvidia Tesla P100 GPU.
    NvidiaTeslaV100
    NVIDIA_TESLA_V100Nvidia V100 GPU.
    NvidiaTeslaP4
    NVIDIA_TESLA_P4Nvidia Tesla P4 GPU.
    NvidiaTeslaT4
    NVIDIA_TESLA_T4Nvidia T4 GPU.
    NvidiaTeslaA100
    NVIDIA_TESLA_A100Nvidia A100 GPU.
    TpuV2
    TPU_V2TPU v2.
    TpuV3
    TPU_V3TPU v3.
    TpuV2Pod
    TPU_V2_PODTPU v2 POD.
    TpuV3Pod
    TPU_V3_PODTPU v3 POD.
    TpuV4Pod
    TPU_V4_PODTPU v4 POD.
    AcceleratorTypeUnspecified
    ACCELERATOR_TYPE_UNSPECIFIEDUnspecified accelerator type. Default to no GPU.
    NvidiaTeslaK80
    NVIDIA_TESLA_K80Nvidia Tesla K80 GPU.
    NvidiaTeslaP100
    NVIDIA_TESLA_P100Nvidia Tesla P100 GPU.
    NvidiaTeslaV100
    NVIDIA_TESLA_V100Nvidia V100 GPU.
    NvidiaTeslaP4
    NVIDIA_TESLA_P4Nvidia Tesla P4 GPU.
    NvidiaTeslaT4
    NVIDIA_TESLA_T4Nvidia T4 GPU.
    NvidiaTeslaA100
    NVIDIA_TESLA_A100Nvidia A100 GPU.
    TpuV2
    TPU_V2TPU v2.
    TpuV3
    TPU_V3TPU v3.
    TpuV2Pod
    TPU_V2_PODTPU v2 POD.
    TpuV3Pod
    TPU_V3_PODTPU v3 POD.
    TpuV4Pod
    TPU_V4_PODTPU v4 POD.
    ACCELERATOR_TYPE_UNSPECIFIED
    ACCELERATOR_TYPE_UNSPECIFIEDUnspecified accelerator type. Default to no GPU.
    NVIDIA_TESLA_K80
    NVIDIA_TESLA_K80Nvidia Tesla K80 GPU.
    NVIDIA_TESLA_P100
    NVIDIA_TESLA_P100Nvidia Tesla P100 GPU.
    NVIDIA_TESLA_V100
    NVIDIA_TESLA_V100Nvidia V100 GPU.
    NVIDIA_TESLA_P4
    NVIDIA_TESLA_P4Nvidia Tesla P4 GPU.
    NVIDIA_TESLA_T4
    NVIDIA_TESLA_T4Nvidia T4 GPU.
    NVIDIA_TESLA_A100
    NVIDIA_TESLA_A100Nvidia A100 GPU.
    TPU_V2
    TPU_V2TPU v2.
    TPU_V3
    TPU_V3TPU v3.
    TPU_V2_POD
    TPU_V2_PODTPU v2 POD.
    TPU_V3_POD
    TPU_V3_PODTPU v3 POD.
    TPU_V4_POD
    TPU_V4_PODTPU v4 POD.
    "ACCELERATOR_TYPE_UNSPECIFIED"
    ACCELERATOR_TYPE_UNSPECIFIEDUnspecified accelerator type. Default to no GPU.
    "NVIDIA_TESLA_K80"
    NVIDIA_TESLA_K80Nvidia Tesla K80 GPU.
    "NVIDIA_TESLA_P100"
    NVIDIA_TESLA_P100Nvidia Tesla P100 GPU.
    "NVIDIA_TESLA_V100"
    NVIDIA_TESLA_V100Nvidia V100 GPU.
    "NVIDIA_TESLA_P4"
    NVIDIA_TESLA_P4Nvidia Tesla P4 GPU.
    "NVIDIA_TESLA_T4"
    NVIDIA_TESLA_T4Nvidia T4 GPU.
    "NVIDIA_TESLA_A100"
    NVIDIA_TESLA_A100Nvidia A100 GPU.
    "TPU_V2"
    TPU_V2TPU v2.
    "TPU_V3"
    TPU_V3TPU v3.
    "TPU_V2_POD"
    TPU_V2_PODTPU v2 POD.
    "TPU_V3_POD"
    TPU_V3_PODTPU v3 POD.
    "TPU_V4_POD"
    TPU_V4_PODTPU v4 POD.

    GoogleCloudMlV1__BuiltInAlgorithmOutput, GoogleCloudMlV1__BuiltInAlgorithmOutputArgs

    Framework string
    Framework on which the built-in algorithm was trained.
    ModelPath string
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    PythonVersion string
    Python version on which the built-in algorithm was trained.
    RuntimeVersion string
    AI Platform runtime version on which the built-in algorithm was trained.
    Framework string
    Framework on which the built-in algorithm was trained.
    ModelPath string
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    PythonVersion string
    Python version on which the built-in algorithm was trained.
    RuntimeVersion string
    AI Platform runtime version on which the built-in algorithm was trained.
    framework String
    Framework on which the built-in algorithm was trained.
    modelPath String
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    pythonVersion String
    Python version on which the built-in algorithm was trained.
    runtimeVersion String
    AI Platform runtime version on which the built-in algorithm was trained.
    framework string
    Framework on which the built-in algorithm was trained.
    modelPath string
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    pythonVersion string
    Python version on which the built-in algorithm was trained.
    runtimeVersion string
    AI Platform runtime version on which the built-in algorithm was trained.
    framework str
    Framework on which the built-in algorithm was trained.
    model_path str
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    python_version str
    Python version on which the built-in algorithm was trained.
    runtime_version str
    AI Platform runtime version on which the built-in algorithm was trained.
    framework String
    Framework on which the built-in algorithm was trained.
    modelPath String
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    pythonVersion String
    Python version on which the built-in algorithm was trained.
    runtimeVersion String
    AI Platform runtime version on which the built-in algorithm was trained.

    GoogleCloudMlV1__BuiltInAlgorithmOutputResponse, GoogleCloudMlV1__BuiltInAlgorithmOutputResponseArgs

    Framework string
    Framework on which the built-in algorithm was trained.
    ModelPath string
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    PythonVersion string
    Python version on which the built-in algorithm was trained.
    RuntimeVersion string
    AI Platform runtime version on which the built-in algorithm was trained.
    Framework string
    Framework on which the built-in algorithm was trained.
    ModelPath string
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    PythonVersion string
    Python version on which the built-in algorithm was trained.
    RuntimeVersion string
    AI Platform runtime version on which the built-in algorithm was trained.
    framework String
    Framework on which the built-in algorithm was trained.
    modelPath String
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    pythonVersion String
    Python version on which the built-in algorithm was trained.
    runtimeVersion String
    AI Platform runtime version on which the built-in algorithm was trained.
    framework string
    Framework on which the built-in algorithm was trained.
    modelPath string
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    pythonVersion string
    Python version on which the built-in algorithm was trained.
    runtimeVersion string
    AI Platform runtime version on which the built-in algorithm was trained.
    framework str
    Framework on which the built-in algorithm was trained.
    model_path str
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    python_version str
    Python version on which the built-in algorithm was trained.
    runtime_version str
    AI Platform runtime version on which the built-in algorithm was trained.
    framework String
    Framework on which the built-in algorithm was trained.
    modelPath String
    The Cloud Storage path to the model/ directory where the training job saves the trained model. Only set for successful jobs that don't use hyperparameter tuning.
    pythonVersion String
    Python version on which the built-in algorithm was trained.
    runtimeVersion String
    AI Platform runtime version on which the built-in algorithm was trained.

    GoogleCloudMlV1__DiskConfig, GoogleCloudMlV1__DiskConfigArgs

    BootDiskSizeGb int
    Size in GB of the boot disk (default is 100GB).
    BootDiskType string
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    BootDiskSizeGb int
    Size in GB of the boot disk (default is 100GB).
    BootDiskType string
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    bootDiskSizeGb Integer
    Size in GB of the boot disk (default is 100GB).
    bootDiskType String
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    bootDiskSizeGb number
    Size in GB of the boot disk (default is 100GB).
    bootDiskType string
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    boot_disk_size_gb int
    Size in GB of the boot disk (default is 100GB).
    boot_disk_type str
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    bootDiskSizeGb Number
    Size in GB of the boot disk (default is 100GB).
    bootDiskType String
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).

    GoogleCloudMlV1__DiskConfigResponse, GoogleCloudMlV1__DiskConfigResponseArgs

    BootDiskSizeGb int
    Size in GB of the boot disk (default is 100GB).
    BootDiskType string
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    BootDiskSizeGb int
    Size in GB of the boot disk (default is 100GB).
    BootDiskType string
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    bootDiskSizeGb Integer
    Size in GB of the boot disk (default is 100GB).
    bootDiskType String
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    bootDiskSizeGb number
    Size in GB of the boot disk (default is 100GB).
    bootDiskType string
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    boot_disk_size_gb int
    Size in GB of the boot disk (default is 100GB).
    boot_disk_type str
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).
    bootDiskSizeGb Number
    Size in GB of the boot disk (default is 100GB).
    bootDiskType String
    Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive).

    GoogleCloudMlV1__EncryptionConfig, GoogleCloudMlV1__EncryptionConfigArgs

    KmsKeyName string
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    KmsKeyName string
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    kmsKeyName String
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    kmsKeyName string
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    kms_key_name str
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    kmsKeyName String
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}

    GoogleCloudMlV1__EncryptionConfigResponse, GoogleCloudMlV1__EncryptionConfigResponseArgs

    KmsKeyName string
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    KmsKeyName string
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    kmsKeyName String
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    kmsKeyName string
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    kms_key_name str
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}
    kmsKeyName String
    The Cloud KMS resource identifier of the customer-managed encryption key used to protect a resource, such as a training job. It has the following format: projects/{PROJECT_ID}/locations/{REGION}/keyRings/{KEY_RING_NAME}/cryptoKeys/{KEY_NAME}

    GoogleCloudMlV1__HyperparameterOutput, GoogleCloudMlV1__HyperparameterOutputArgs

    AllMetrics List<Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetric>
    All recorded object metrics for this trial. This field is not currently populated.
    BuiltInAlgorithmOutput Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    FinalMetric Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetric
    The final objective metric seen for this trial.
    Hyperparameters Dictionary<string, string>
    The hyperparameters given to this trial.
    IsTrialStoppedEarly bool
    True if the trial is stopped early.
    TrialId string
    The trial id for these results.
    WebAccessUris Dictionary<string, string>
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    AllMetrics []GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetric
    All recorded object metrics for this trial. This field is not currently populated.
    BuiltInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    FinalMetric GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetric
    The final objective metric seen for this trial.
    Hyperparameters map[string]string
    The hyperparameters given to this trial.
    IsTrialStoppedEarly bool
    True if the trial is stopped early.
    TrialId string
    The trial id for these results.
    WebAccessUris map[string]string
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    allMetrics List<GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetric>
    All recorded object metrics for this trial. This field is not currently populated.
    builtInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    finalMetric GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetric
    The final objective metric seen for this trial.
    hyperparameters Map<String,String>
    The hyperparameters given to this trial.
    isTrialStoppedEarly Boolean
    True if the trial is stopped early.
    trialId String
    The trial id for these results.
    webAccessUris Map<String,String>
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    allMetrics GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetric[]
    All recorded object metrics for this trial. This field is not currently populated.
    builtInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    finalMetric GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetric
    The final objective metric seen for this trial.
    hyperparameters {[key: string]: string}
    The hyperparameters given to this trial.
    isTrialStoppedEarly boolean
    True if the trial is stopped early.
    trialId string
    The trial id for these results.
    webAccessUris {[key: string]: string}
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    all_metrics Sequence[GoogleCloudMlV1HyperparameterOutput_HyperparameterMetric]
    All recorded object metrics for this trial. This field is not currently populated.
    built_in_algorithm_output GoogleCloudMlV1BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    final_metric GoogleCloudMlV1HyperparameterOutput_HyperparameterMetric
    The final objective metric seen for this trial.
    hyperparameters Mapping[str, str]
    The hyperparameters given to this trial.
    is_trial_stopped_early bool
    True if the trial is stopped early.
    trial_id str
    The trial id for these results.
    web_access_uris Mapping[str, str]
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    allMetrics List<Property Map>
    All recorded object metrics for this trial. This field is not currently populated.
    builtInAlgorithmOutput Property Map
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    finalMetric Property Map
    The final objective metric seen for this trial.
    hyperparameters Map<String>
    The hyperparameters given to this trial.
    isTrialStoppedEarly Boolean
    True if the trial is stopped early.
    trialId String
    The trial id for these results.
    webAccessUris Map<String>
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.

    GoogleCloudMlV1__HyperparameterOutputResponse, GoogleCloudMlV1__HyperparameterOutputResponseArgs

    AllMetrics List<Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponse>
    All recorded object metrics for this trial. This field is not currently populated.
    BuiltInAlgorithmOutput Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    EndTime string
    End time for the trial.
    FinalMetric Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponse
    The final objective metric seen for this trial.
    Hyperparameters Dictionary<string, string>
    The hyperparameters given to this trial.
    IsTrialStoppedEarly bool
    True if the trial is stopped early.
    StartTime string
    Start time for the trial.
    State string
    The detailed state of the trial.
    TrialId string
    The trial id for these results.
    WebAccessUris Dictionary<string, string>
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    AllMetrics []GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponse
    All recorded object metrics for this trial. This field is not currently populated.
    BuiltInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    EndTime string
    End time for the trial.
    FinalMetric GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponse
    The final objective metric seen for this trial.
    Hyperparameters map[string]string
    The hyperparameters given to this trial.
    IsTrialStoppedEarly bool
    True if the trial is stopped early.
    StartTime string
    Start time for the trial.
    State string
    The detailed state of the trial.
    TrialId string
    The trial id for these results.
    WebAccessUris map[string]string
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    allMetrics List<GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponse>
    All recorded object metrics for this trial. This field is not currently populated.
    builtInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    endTime String
    End time for the trial.
    finalMetric GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponse
    The final objective metric seen for this trial.
    hyperparameters Map<String,String>
    The hyperparameters given to this trial.
    isTrialStoppedEarly Boolean
    True if the trial is stopped early.
    startTime String
    Start time for the trial.
    state String
    The detailed state of the trial.
    trialId String
    The trial id for these results.
    webAccessUris Map<String,String>
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    allMetrics GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponse[]
    All recorded object metrics for this trial. This field is not currently populated.
    builtInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    endTime string
    End time for the trial.
    finalMetric GoogleCloudMlV1_HyperparameterOutput_HyperparameterMetricResponse
    The final objective metric seen for this trial.
    hyperparameters {[key: string]: string}
    The hyperparameters given to this trial.
    isTrialStoppedEarly boolean
    True if the trial is stopped early.
    startTime string
    Start time for the trial.
    state string
    The detailed state of the trial.
    trialId string
    The trial id for these results.
    webAccessUris {[key: string]: string}
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    all_metrics Sequence[GoogleCloudMlV1HyperparameterOutput_HyperparameterMetricResponse]
    All recorded object metrics for this trial. This field is not currently populated.
    built_in_algorithm_output GoogleCloudMlV1BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    end_time str
    End time for the trial.
    final_metric GoogleCloudMlV1HyperparameterOutput_HyperparameterMetricResponse
    The final objective metric seen for this trial.
    hyperparameters Mapping[str, str]
    The hyperparameters given to this trial.
    is_trial_stopped_early bool
    True if the trial is stopped early.
    start_time str
    Start time for the trial.
    state str
    The detailed state of the trial.
    trial_id str
    The trial id for these results.
    web_access_uris Mapping[str, str]
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    allMetrics List<Property Map>
    All recorded object metrics for this trial. This field is not currently populated.
    builtInAlgorithmOutput Property Map
    Details related to built-in algorithms jobs. Only set for trials of built-in algorithms jobs that have succeeded.
    endTime String
    End time for the trial.
    finalMetric Property Map
    The final objective metric seen for this trial.
    hyperparameters Map<String>
    The hyperparameters given to this trial.
    isTrialStoppedEarly Boolean
    True if the trial is stopped early.
    startTime String
    Start time for the trial.
    state String
    The detailed state of the trial.
    trialId String
    The trial id for these results.
    webAccessUris Map<String>
    URIs for accessing interactive shells (one URI for each training node). Only available if this trial is part of a hyperparameter tuning job and the job's training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.

    GoogleCloudMlV1__HyperparameterSpec, GoogleCloudMlV1__HyperparameterSpecArgs

    Goal Pulumi.GoogleNative.Ml.V1.GoogleCloudMlV1__HyperparameterSpecGoal
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    Params List<Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ParameterSpec>
    The set of parameters to tune.
    Algorithm Pulumi.GoogleNative.Ml.V1.GoogleCloudMlV1__HyperparameterSpecAlgorithm
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    EnableTrialEarlyStopping bool
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    HyperparameterMetricTag string
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    MaxFailedTrials int
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    MaxParallelTrials int
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    MaxTrials int
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    ResumePreviousJobId string
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    Goal GoogleCloudMlV1__HyperparameterSpecGoal
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    Params []GoogleCloudMlV1__ParameterSpec
    The set of parameters to tune.
    Algorithm GoogleCloudMlV1__HyperparameterSpecAlgorithm
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    EnableTrialEarlyStopping bool
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    HyperparameterMetricTag string
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    MaxFailedTrials int
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    MaxParallelTrials int
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    MaxTrials int
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    ResumePreviousJobId string
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    goal GoogleCloudMlV1__HyperparameterSpecGoal
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    params List<GoogleCloudMlV1__ParameterSpec>
    The set of parameters to tune.
    algorithm GoogleCloudMlV1__HyperparameterSpecAlgorithm
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    enableTrialEarlyStopping Boolean
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    hyperparameterMetricTag String
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    maxFailedTrials Integer
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    maxParallelTrials Integer
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    maxTrials Integer
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    resumePreviousJobId String
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    goal GoogleCloudMlV1__HyperparameterSpecGoal
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    params GoogleCloudMlV1__ParameterSpec[]
    The set of parameters to tune.
    algorithm GoogleCloudMlV1__HyperparameterSpecAlgorithm
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    enableTrialEarlyStopping boolean
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    hyperparameterMetricTag string
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    maxFailedTrials number
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    maxParallelTrials number
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    maxTrials number
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    resumePreviousJobId string
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    goal GoogleCloudMlV1HyperparameterSpecGoal
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    params Sequence[GoogleCloudMlV1ParameterSpec]
    The set of parameters to tune.
    algorithm GoogleCloudMlV1HyperparameterSpecAlgorithm
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    enable_trial_early_stopping bool
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    hyperparameter_metric_tag str
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    max_failed_trials int
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    max_parallel_trials int
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    max_trials int
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    resume_previous_job_id str
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    goal "GOAL_TYPE_UNSPECIFIED" | "MAXIMIZE" | "MINIMIZE"
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    params List<Property Map>
    The set of parameters to tune.
    algorithm "ALGORITHM_UNSPECIFIED" | "GRID_SEARCH" | "RANDOM_SEARCH"
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    enableTrialEarlyStopping Boolean
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    hyperparameterMetricTag String
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    maxFailedTrials Number
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    maxParallelTrials Number
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    maxTrials Number
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    resumePreviousJobId String
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.

    GoogleCloudMlV1__HyperparameterSpecAlgorithm, GoogleCloudMlV1__HyperparameterSpecAlgorithmArgs

    AlgorithmUnspecified
    ALGORITHM_UNSPECIFIEDThe default algorithm used by the hyperparameter tuning service. This is a Bayesian optimization algorithm.
    GridSearch
    GRID_SEARCHSimple grid search within the feasible space. To use grid search, all parameters must be INTEGER, CATEGORICAL, or DISCRETE.
    RandomSearch
    RANDOM_SEARCHSimple random search within the feasible space.
    GoogleCloudMlV1__HyperparameterSpecAlgorithmAlgorithmUnspecified
    ALGORITHM_UNSPECIFIEDThe default algorithm used by the hyperparameter tuning service. This is a Bayesian optimization algorithm.
    GoogleCloudMlV1__HyperparameterSpecAlgorithmGridSearch
    GRID_SEARCHSimple grid search within the feasible space. To use grid search, all parameters must be INTEGER, CATEGORICAL, or DISCRETE.
    GoogleCloudMlV1__HyperparameterSpecAlgorithmRandomSearch
    RANDOM_SEARCHSimple random search within the feasible space.
    AlgorithmUnspecified
    ALGORITHM_UNSPECIFIEDThe default algorithm used by the hyperparameter tuning service. This is a Bayesian optimization algorithm.
    GridSearch
    GRID_SEARCHSimple grid search within the feasible space. To use grid search, all parameters must be INTEGER, CATEGORICAL, or DISCRETE.
    RandomSearch
    RANDOM_SEARCHSimple random search within the feasible space.
    AlgorithmUnspecified
    ALGORITHM_UNSPECIFIEDThe default algorithm used by the hyperparameter tuning service. This is a Bayesian optimization algorithm.
    GridSearch
    GRID_SEARCHSimple grid search within the feasible space. To use grid search, all parameters must be INTEGER, CATEGORICAL, or DISCRETE.
    RandomSearch
    RANDOM_SEARCHSimple random search within the feasible space.
    ALGORITHM_UNSPECIFIED
    ALGORITHM_UNSPECIFIEDThe default algorithm used by the hyperparameter tuning service. This is a Bayesian optimization algorithm.
    GRID_SEARCH
    GRID_SEARCHSimple grid search within the feasible space. To use grid search, all parameters must be INTEGER, CATEGORICAL, or DISCRETE.
    RANDOM_SEARCH
    RANDOM_SEARCHSimple random search within the feasible space.
    "ALGORITHM_UNSPECIFIED"
    ALGORITHM_UNSPECIFIEDThe default algorithm used by the hyperparameter tuning service. This is a Bayesian optimization algorithm.
    "GRID_SEARCH"
    GRID_SEARCHSimple grid search within the feasible space. To use grid search, all parameters must be INTEGER, CATEGORICAL, or DISCRETE.
    "RANDOM_SEARCH"
    RANDOM_SEARCHSimple random search within the feasible space.

    GoogleCloudMlV1__HyperparameterSpecGoal, GoogleCloudMlV1__HyperparameterSpecGoalArgs

    GoalTypeUnspecified
    GOAL_TYPE_UNSPECIFIEDGoal Type will default to maximize.
    Maximize
    MAXIMIZEMaximize the goal metric.
    Minimize
    MINIMIZEMinimize the goal metric.
    GoogleCloudMlV1__HyperparameterSpecGoalGoalTypeUnspecified
    GOAL_TYPE_UNSPECIFIEDGoal Type will default to maximize.
    GoogleCloudMlV1__HyperparameterSpecGoalMaximize
    MAXIMIZEMaximize the goal metric.
    GoogleCloudMlV1__HyperparameterSpecGoalMinimize
    MINIMIZEMinimize the goal metric.
    GoalTypeUnspecified
    GOAL_TYPE_UNSPECIFIEDGoal Type will default to maximize.
    Maximize
    MAXIMIZEMaximize the goal metric.
    Minimize
    MINIMIZEMinimize the goal metric.
    GoalTypeUnspecified
    GOAL_TYPE_UNSPECIFIEDGoal Type will default to maximize.
    Maximize
    MAXIMIZEMaximize the goal metric.
    Minimize
    MINIMIZEMinimize the goal metric.
    GOAL_TYPE_UNSPECIFIED
    GOAL_TYPE_UNSPECIFIEDGoal Type will default to maximize.
    MAXIMIZE
    MAXIMIZEMaximize the goal metric.
    MINIMIZE
    MINIMIZEMinimize the goal metric.
    "GOAL_TYPE_UNSPECIFIED"
    GOAL_TYPE_UNSPECIFIEDGoal Type will default to maximize.
    "MAXIMIZE"
    MAXIMIZEMaximize the goal metric.
    "MINIMIZE"
    MINIMIZEMinimize the goal metric.

    GoogleCloudMlV1__HyperparameterSpecResponse, GoogleCloudMlV1__HyperparameterSpecResponseArgs

    Algorithm string
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    EnableTrialEarlyStopping bool
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    Goal string
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    HyperparameterMetricTag string
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    MaxFailedTrials int
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    MaxParallelTrials int
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    MaxTrials int
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    Params List<Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ParameterSpecResponse>
    The set of parameters to tune.
    ResumePreviousJobId string
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    Algorithm string
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    EnableTrialEarlyStopping bool
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    Goal string
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    HyperparameterMetricTag string
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    MaxFailedTrials int
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    MaxParallelTrials int
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    MaxTrials int
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    Params []GoogleCloudMlV1__ParameterSpecResponse
    The set of parameters to tune.
    ResumePreviousJobId string
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    algorithm String
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    enableTrialEarlyStopping Boolean
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    goal String
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    hyperparameterMetricTag String
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    maxFailedTrials Integer
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    maxParallelTrials Integer
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    maxTrials Integer
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    params List<GoogleCloudMlV1__ParameterSpecResponse>
    The set of parameters to tune.
    resumePreviousJobId String
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    algorithm string
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    enableTrialEarlyStopping boolean
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    goal string
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    hyperparameterMetricTag string
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    maxFailedTrials number
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    maxParallelTrials number
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    maxTrials number
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    params GoogleCloudMlV1__ParameterSpecResponse[]
    The set of parameters to tune.
    resumePreviousJobId string
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    algorithm str
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    enable_trial_early_stopping bool
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    goal str
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    hyperparameter_metric_tag str
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    max_failed_trials int
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    max_parallel_trials int
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    max_trials int
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    params Sequence[GoogleCloudMlV1ParameterSpecResponse]
    The set of parameters to tune.
    resume_previous_job_id str
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.
    algorithm String
    Optional. The search algorithm specified for the hyperparameter tuning job. Uses the default AI Platform hyperparameter tuning algorithm if unspecified.
    enableTrialEarlyStopping Boolean
    Optional. Indicates if the hyperparameter tuning job enables auto trial early stopping.
    goal String
    The type of goal to use for tuning. Available types are MAXIMIZE and MINIMIZE. Defaults to MAXIMIZE.
    hyperparameterMetricTag String
    Optional. The TensorFlow summary tag name to use for optimizing trials. For current versions of TensorFlow, this tag name should exactly match what is shown in TensorBoard, including all scopes. For versions of TensorFlow prior to 0.12, this should be only the tag passed to tf.Summary. By default, "training/hptuning/metric" will be used.
    maxFailedTrials Number
    Optional. The number of failed trials that need to be seen before failing the hyperparameter tuning job. You can specify this field to override the default failing criteria for AI Platform hyperparameter tuning jobs. Defaults to zero, which means the service decides when a hyperparameter job should fail.
    maxParallelTrials Number
    Optional. The number of training trials to run concurrently. You can reduce the time it takes to perform hyperparameter tuning by adding trials in parallel. However, each trail only benefits from the information gained in completed trials. That means that a trial does not get access to the results of trials running at the same time, which could reduce the quality of the overall optimization. Each trial will use the same scale tier and machine types. Defaults to one.
    maxTrials Number
    Optional. How many training trials should be attempted to optimize the specified hyperparameters. Defaults to one.
    params List<Property Map>
    The set of parameters to tune.
    resumePreviousJobId String
    Optional. The prior hyperparameter tuning job id that users hope to continue with. The job id will be used to find the corresponding vizier study guid and resume the study.

    GoogleCloudMlV1__ParameterSpec, GoogleCloudMlV1__ParameterSpecArgs

    ParameterName string
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    Type Pulumi.GoogleNative.Ml.V1.GoogleCloudMlV1__ParameterSpecType
    The type of the parameter.
    CategoricalValues List<string>
    Required if type is CATEGORICAL. The list of possible categories.
    DiscreteValues List<double>
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    MaxValue double
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    MinValue double
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    ScaleType Pulumi.GoogleNative.Ml.V1.GoogleCloudMlV1__ParameterSpecScaleType
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    ParameterName string
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    Type GoogleCloudMlV1__ParameterSpecType
    The type of the parameter.
    CategoricalValues []string
    Required if type is CATEGORICAL. The list of possible categories.
    DiscreteValues []float64
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    MaxValue float64
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    MinValue float64
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    ScaleType GoogleCloudMlV1__ParameterSpecScaleType
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    parameterName String
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    type GoogleCloudMlV1__ParameterSpecType
    The type of the parameter.
    categoricalValues List<String>
    Required if type is CATEGORICAL. The list of possible categories.
    discreteValues List<Double>
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    maxValue Double
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    minValue Double
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    scaleType GoogleCloudMlV1__ParameterSpecScaleType
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    parameterName string
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    type GoogleCloudMlV1__ParameterSpecType
    The type of the parameter.
    categoricalValues string[]
    Required if type is CATEGORICAL. The list of possible categories.
    discreteValues number[]
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    maxValue number
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    minValue number
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    scaleType GoogleCloudMlV1__ParameterSpecScaleType
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    parameter_name str
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    type GoogleCloudMlV1ParameterSpecType
    The type of the parameter.
    categorical_values Sequence[str]
    Required if type is CATEGORICAL. The list of possible categories.
    discrete_values Sequence[float]
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    max_value float
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    min_value float
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    scale_type GoogleCloudMlV1ParameterSpecScaleType
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    parameterName String
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    type "PARAMETER_TYPE_UNSPECIFIED" | "DOUBLE" | "INTEGER" | "CATEGORICAL" | "DISCRETE"
    The type of the parameter.
    categoricalValues List<String>
    Required if type is CATEGORICAL. The list of possible categories.
    discreteValues List<Number>
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    maxValue Number
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    minValue Number
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    scaleType "NONE" | "UNIT_LINEAR_SCALE" | "UNIT_LOG_SCALE" | "UNIT_REVERSE_LOG_SCALE"
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).

    GoogleCloudMlV1__ParameterSpecResponse, GoogleCloudMlV1__ParameterSpecResponseArgs

    CategoricalValues List<string>
    Required if type is CATEGORICAL. The list of possible categories.
    DiscreteValues List<double>
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    MaxValue double
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    MinValue double
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    ParameterName string
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    ScaleType string
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    Type string
    The type of the parameter.
    CategoricalValues []string
    Required if type is CATEGORICAL. The list of possible categories.
    DiscreteValues []float64
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    MaxValue float64
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    MinValue float64
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    ParameterName string
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    ScaleType string
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    Type string
    The type of the parameter.
    categoricalValues List<String>
    Required if type is CATEGORICAL. The list of possible categories.
    discreteValues List<Double>
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    maxValue Double
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    minValue Double
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    parameterName String
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    scaleType String
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    type String
    The type of the parameter.
    categoricalValues string[]
    Required if type is CATEGORICAL. The list of possible categories.
    discreteValues number[]
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    maxValue number
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    minValue number
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    parameterName string
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    scaleType string
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    type string
    The type of the parameter.
    categorical_values Sequence[str]
    Required if type is CATEGORICAL. The list of possible categories.
    discrete_values Sequence[float]
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    max_value float
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    min_value float
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    parameter_name str
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    scale_type str
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    type str
    The type of the parameter.
    categoricalValues List<String>
    Required if type is CATEGORICAL. The list of possible categories.
    discreteValues List<Number>
    Required if type is DISCRETE. A list of feasible points. The list should be in strictly increasing order. For instance, this parameter might have possible settings of 1.5, 2.5, and 4.0. This list should not contain more than 1,000 values.
    maxValue Number
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    minValue Number
    Required if type is DOUBLE or INTEGER. This field should be unset if type is CATEGORICAL. This value should be integers if type is INTEGER.
    parameterName String
    The parameter name must be unique amongst all ParameterConfigs in a HyperparameterSpec message. E.g., "learning_rate".
    scaleType String
    Optional. How the parameter should be scaled to the hypercube. Leave unset for categorical parameters. Some kind of scaling is strongly recommended for real or integral parameters (e.g., UNIT_LINEAR_SCALE).
    type String
    The type of the parameter.

    GoogleCloudMlV1__ParameterSpecScaleType, GoogleCloudMlV1__ParameterSpecScaleTypeArgs

    None
    NONEBy default, no scaling is applied.
    UnitLinearScale
    UNIT_LINEAR_SCALEScales the feasible space to (0, 1) linearly.
    UnitLogScale
    UNIT_LOG_SCALEScales the feasible space logarithmically to (0, 1). The entire feasible space must be strictly positive.
    UnitReverseLogScale
    UNIT_REVERSE_LOG_SCALEScales the feasible space "reverse" logarithmically to (0, 1). The result is that values close to the top of the feasible space are spread out more than points near the bottom. The entire feasible space must be strictly positive.
    GoogleCloudMlV1__ParameterSpecScaleTypeNone
    NONEBy default, no scaling is applied.
    GoogleCloudMlV1__ParameterSpecScaleTypeUnitLinearScale
    UNIT_LINEAR_SCALEScales the feasible space to (0, 1) linearly.
    GoogleCloudMlV1__ParameterSpecScaleTypeUnitLogScale
    UNIT_LOG_SCALEScales the feasible space logarithmically to (0, 1). The entire feasible space must be strictly positive.
    GoogleCloudMlV1__ParameterSpecScaleTypeUnitReverseLogScale
    UNIT_REVERSE_LOG_SCALEScales the feasible space "reverse" logarithmically to (0, 1). The result is that values close to the top of the feasible space are spread out more than points near the bottom. The entire feasible space must be strictly positive.
    None
    NONEBy default, no scaling is applied.
    UnitLinearScale
    UNIT_LINEAR_SCALEScales the feasible space to (0, 1) linearly.
    UnitLogScale
    UNIT_LOG_SCALEScales the feasible space logarithmically to (0, 1). The entire feasible space must be strictly positive.
    UnitReverseLogScale
    UNIT_REVERSE_LOG_SCALEScales the feasible space "reverse" logarithmically to (0, 1). The result is that values close to the top of the feasible space are spread out more than points near the bottom. The entire feasible space must be strictly positive.
    None
    NONEBy default, no scaling is applied.
    UnitLinearScale
    UNIT_LINEAR_SCALEScales the feasible space to (0, 1) linearly.
    UnitLogScale
    UNIT_LOG_SCALEScales the feasible space logarithmically to (0, 1). The entire feasible space must be strictly positive.
    UnitReverseLogScale
    UNIT_REVERSE_LOG_SCALEScales the feasible space "reverse" logarithmically to (0, 1). The result is that values close to the top of the feasible space are spread out more than points near the bottom. The entire feasible space must be strictly positive.
    NONE
    NONEBy default, no scaling is applied.
    UNIT_LINEAR_SCALE
    UNIT_LINEAR_SCALEScales the feasible space to (0, 1) linearly.
    UNIT_LOG_SCALE
    UNIT_LOG_SCALEScales the feasible space logarithmically to (0, 1). The entire feasible space must be strictly positive.
    UNIT_REVERSE_LOG_SCALE
    UNIT_REVERSE_LOG_SCALEScales the feasible space "reverse" logarithmically to (0, 1). The result is that values close to the top of the feasible space are spread out more than points near the bottom. The entire feasible space must be strictly positive.
    "NONE"
    NONEBy default, no scaling is applied.
    "UNIT_LINEAR_SCALE"
    UNIT_LINEAR_SCALEScales the feasible space to (0, 1) linearly.
    "UNIT_LOG_SCALE"
    UNIT_LOG_SCALEScales the feasible space logarithmically to (0, 1). The entire feasible space must be strictly positive.
    "UNIT_REVERSE_LOG_SCALE"
    UNIT_REVERSE_LOG_SCALEScales the feasible space "reverse" logarithmically to (0, 1). The result is that values close to the top of the feasible space are spread out more than points near the bottom. The entire feasible space must be strictly positive.

    GoogleCloudMlV1__ParameterSpecType, GoogleCloudMlV1__ParameterSpecTypeArgs

    ParameterTypeUnspecified
    PARAMETER_TYPE_UNSPECIFIEDYou must specify a valid type. Using this unspecified type will result in an error.
    Double
    DOUBLEType for real-valued parameters.
    Integer
    INTEGERType for integral parameters.
    Categorical
    CATEGORICALThe parameter is categorical, with a value chosen from the categories field.
    Discrete
    DISCRETEThe parameter is real valued, with a fixed set of feasible points. If type==DISCRETE, feasible_points must be provided, and {min_value, max_value} will be ignored.
    GoogleCloudMlV1__ParameterSpecTypeParameterTypeUnspecified
    PARAMETER_TYPE_UNSPECIFIEDYou must specify a valid type. Using this unspecified type will result in an error.
    GoogleCloudMlV1__ParameterSpecTypeDouble
    DOUBLEType for real-valued parameters.
    GoogleCloudMlV1__ParameterSpecTypeInteger
    INTEGERType for integral parameters.
    GoogleCloudMlV1__ParameterSpecTypeCategorical
    CATEGORICALThe parameter is categorical, with a value chosen from the categories field.
    GoogleCloudMlV1__ParameterSpecTypeDiscrete
    DISCRETEThe parameter is real valued, with a fixed set of feasible points. If type==DISCRETE, feasible_points must be provided, and {min_value, max_value} will be ignored.
    ParameterTypeUnspecified
    PARAMETER_TYPE_UNSPECIFIEDYou must specify a valid type. Using this unspecified type will result in an error.
    Double
    DOUBLEType for real-valued parameters.
    Integer
    INTEGERType for integral parameters.
    Categorical
    CATEGORICALThe parameter is categorical, with a value chosen from the categories field.
    Discrete
    DISCRETEThe parameter is real valued, with a fixed set of feasible points. If type==DISCRETE, feasible_points must be provided, and {min_value, max_value} will be ignored.
    ParameterTypeUnspecified
    PARAMETER_TYPE_UNSPECIFIEDYou must specify a valid type. Using this unspecified type will result in an error.
    Double
    DOUBLEType for real-valued parameters.
    Integer
    INTEGERType for integral parameters.
    Categorical
    CATEGORICALThe parameter is categorical, with a value chosen from the categories field.
    Discrete
    DISCRETEThe parameter is real valued, with a fixed set of feasible points. If type==DISCRETE, feasible_points must be provided, and {min_value, max_value} will be ignored.
    PARAMETER_TYPE_UNSPECIFIED
    PARAMETER_TYPE_UNSPECIFIEDYou must specify a valid type. Using this unspecified type will result in an error.
    DOUBLE
    DOUBLEType for real-valued parameters.
    INTEGER
    INTEGERType for integral parameters.
    CATEGORICAL
    CATEGORICALThe parameter is categorical, with a value chosen from the categories field.
    DISCRETE
    DISCRETEThe parameter is real valued, with a fixed set of feasible points. If type==DISCRETE, feasible_points must be provided, and {min_value, max_value} will be ignored.
    "PARAMETER_TYPE_UNSPECIFIED"
    PARAMETER_TYPE_UNSPECIFIEDYou must specify a valid type. Using this unspecified type will result in an error.
    "DOUBLE"
    DOUBLEType for real-valued parameters.
    "INTEGER"
    INTEGERType for integral parameters.
    "CATEGORICAL"
    CATEGORICALThe parameter is categorical, with a value chosen from the categories field.
    "DISCRETE"
    DISCRETEThe parameter is real valued, with a fixed set of feasible points. If type==DISCRETE, feasible_points must be provided, and {min_value, max_value} will be ignored.

    GoogleCloudMlV1__PredictionInput, GoogleCloudMlV1__PredictionInputArgs

    DataFormat Pulumi.GoogleNative.Ml.V1.GoogleCloudMlV1__PredictionInputDataFormat
    The format of the input data files.
    InputPaths List<string>
    The Cloud Storage location of the input data files. May contain wildcards.
    OutputPath string
    The output Google Cloud Storage location.
    Region string
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    BatchSize string
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    MaxWorkerCount string
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    ModelName string
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    OutputDataFormat Pulumi.GoogleNative.Ml.V1.GoogleCloudMlV1__PredictionInputOutputDataFormat
    Optional. Format of the output data files, defaults to JSON.
    RuntimeVersion string
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    SignatureName string
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    Uri string
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    VersionName string
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    DataFormat GoogleCloudMlV1__PredictionInputDataFormat
    The format of the input data files.
    InputPaths []string
    The Cloud Storage location of the input data files. May contain wildcards.
    OutputPath string
    The output Google Cloud Storage location.
    Region string
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    BatchSize string
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    MaxWorkerCount string
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    ModelName string
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    OutputDataFormat GoogleCloudMlV1__PredictionInputOutputDataFormat
    Optional. Format of the output data files, defaults to JSON.
    RuntimeVersion string
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    SignatureName string
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    Uri string
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    VersionName string
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    dataFormat GoogleCloudMlV1__PredictionInputDataFormat
    The format of the input data files.
    inputPaths List<String>
    The Cloud Storage location of the input data files. May contain wildcards.
    outputPath String
    The output Google Cloud Storage location.
    region String
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    batchSize String
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    maxWorkerCount String
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    modelName String
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    outputDataFormat GoogleCloudMlV1__PredictionInputOutputDataFormat
    Optional. Format of the output data files, defaults to JSON.
    runtimeVersion String
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    signatureName String
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    uri String
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    versionName String
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    dataFormat GoogleCloudMlV1__PredictionInputDataFormat
    The format of the input data files.
    inputPaths string[]
    The Cloud Storage location of the input data files. May contain wildcards.
    outputPath string
    The output Google Cloud Storage location.
    region string
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    batchSize string
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    maxWorkerCount string
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    modelName string
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    outputDataFormat GoogleCloudMlV1__PredictionInputOutputDataFormat
    Optional. Format of the output data files, defaults to JSON.
    runtimeVersion string
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    signatureName string
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    uri string
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    versionName string
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    data_format GoogleCloudMlV1PredictionInputDataFormat
    The format of the input data files.
    input_paths Sequence[str]
    The Cloud Storage location of the input data files. May contain wildcards.
    output_path str
    The output Google Cloud Storage location.
    region str
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    batch_size str
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    max_worker_count str
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    model_name str
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    output_data_format GoogleCloudMlV1PredictionInputOutputDataFormat
    Optional. Format of the output data files, defaults to JSON.
    runtime_version str
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    signature_name str
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    uri str
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    version_name str
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    dataFormat "DATA_FORMAT_UNSPECIFIED" | "JSON" | "TEXT" | "TF_RECORD" | "TF_RECORD_GZIP" | "CSV"
    The format of the input data files.
    inputPaths List<String>
    The Cloud Storage location of the input data files. May contain wildcards.
    outputPath String
    The output Google Cloud Storage location.
    region String
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    batchSize String
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    maxWorkerCount String
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    modelName String
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    outputDataFormat "DATA_FORMAT_UNSPECIFIED" | "JSON" | "TEXT" | "TF_RECORD" | "TF_RECORD_GZIP" | "CSV"
    Optional. Format of the output data files, defaults to JSON.
    runtimeVersion String
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    signatureName String
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    uri String
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    versionName String
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"

    GoogleCloudMlV1__PredictionInputDataFormat, GoogleCloudMlV1__PredictionInputDataFormatArgs

    DataFormatUnspecified
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    Json
    JSONEach line of the file is a JSON dictionary representing one record.
    Text
    TEXTDeprecated. Use JSON instead.
    TfRecord
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    TfRecordGzip
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    Csv
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    GoogleCloudMlV1__PredictionInputDataFormatDataFormatUnspecified
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    GoogleCloudMlV1__PredictionInputDataFormatJson
    JSONEach line of the file is a JSON dictionary representing one record.
    GoogleCloudMlV1__PredictionInputDataFormatText
    TEXTDeprecated. Use JSON instead.
    GoogleCloudMlV1__PredictionInputDataFormatTfRecord
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    GoogleCloudMlV1__PredictionInputDataFormatTfRecordGzip
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    GoogleCloudMlV1__PredictionInputDataFormatCsv
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    DataFormatUnspecified
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    Json
    JSONEach line of the file is a JSON dictionary representing one record.
    Text
    TEXTDeprecated. Use JSON instead.
    TfRecord
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    TfRecordGzip
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    Csv
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    DataFormatUnspecified
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    Json
    JSONEach line of the file is a JSON dictionary representing one record.
    Text
    TEXTDeprecated. Use JSON instead.
    TfRecord
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    TfRecordGzip
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    Csv
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    DATA_FORMAT_UNSPECIFIED
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    JSON
    JSONEach line of the file is a JSON dictionary representing one record.
    TEXT
    TEXTDeprecated. Use JSON instead.
    TF_RECORD
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    TF_RECORD_GZIP
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    CSV
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    "DATA_FORMAT_UNSPECIFIED"
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    "JSON"
    JSONEach line of the file is a JSON dictionary representing one record.
    "TEXT"
    TEXTDeprecated. Use JSON instead.
    "TF_RECORD"
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    "TF_RECORD_GZIP"
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    "CSV"
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.

    GoogleCloudMlV1__PredictionInputOutputDataFormat, GoogleCloudMlV1__PredictionInputOutputDataFormatArgs

    DataFormatUnspecified
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    Json
    JSONEach line of the file is a JSON dictionary representing one record.
    Text
    TEXTDeprecated. Use JSON instead.
    TfRecord
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    TfRecordGzip
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    Csv
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    GoogleCloudMlV1__PredictionInputOutputDataFormatDataFormatUnspecified
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    GoogleCloudMlV1__PredictionInputOutputDataFormatJson
    JSONEach line of the file is a JSON dictionary representing one record.
    GoogleCloudMlV1__PredictionInputOutputDataFormatText
    TEXTDeprecated. Use JSON instead.
    GoogleCloudMlV1__PredictionInputOutputDataFormatTfRecord
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    GoogleCloudMlV1__PredictionInputOutputDataFormatTfRecordGzip
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    GoogleCloudMlV1__PredictionInputOutputDataFormatCsv
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    DataFormatUnspecified
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    Json
    JSONEach line of the file is a JSON dictionary representing one record.
    Text
    TEXTDeprecated. Use JSON instead.
    TfRecord
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    TfRecordGzip
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    Csv
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    DataFormatUnspecified
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    Json
    JSONEach line of the file is a JSON dictionary representing one record.
    Text
    TEXTDeprecated. Use JSON instead.
    TfRecord
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    TfRecordGzip
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    Csv
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    DATA_FORMAT_UNSPECIFIED
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    JSON
    JSONEach line of the file is a JSON dictionary representing one record.
    TEXT
    TEXTDeprecated. Use JSON instead.
    TF_RECORD
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    TF_RECORD_GZIP
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    CSV
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.
    "DATA_FORMAT_UNSPECIFIED"
    DATA_FORMAT_UNSPECIFIEDUnspecified format.
    "JSON"
    JSONEach line of the file is a JSON dictionary representing one record.
    "TEXT"
    TEXTDeprecated. Use JSON instead.
    "TF_RECORD"
    TF_RECORDThe source file is a TFRecord file. Currently available only for input data.
    "TF_RECORD_GZIP"
    TF_RECORD_GZIPThe source file is a GZIP-compressed TFRecord file. Currently available only for input data.
    "CSV"
    CSVValues are comma-separated rows, with keys in a separate file. Currently available only for output data.

    GoogleCloudMlV1__PredictionInputResponse, GoogleCloudMlV1__PredictionInputResponseArgs

    BatchSize string
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    DataFormat string
    The format of the input data files.
    InputPaths List<string>
    The Cloud Storage location of the input data files. May contain wildcards.
    MaxWorkerCount string
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    ModelName string
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    OutputDataFormat string
    Optional. Format of the output data files, defaults to JSON.
    OutputPath string
    The output Google Cloud Storage location.
    Region string
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    RuntimeVersion string
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    SignatureName string
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    Uri string
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    VersionName string
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    BatchSize string
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    DataFormat string
    The format of the input data files.
    InputPaths []string
    The Cloud Storage location of the input data files. May contain wildcards.
    MaxWorkerCount string
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    ModelName string
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    OutputDataFormat string
    Optional. Format of the output data files, defaults to JSON.
    OutputPath string
    The output Google Cloud Storage location.
    Region string
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    RuntimeVersion string
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    SignatureName string
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    Uri string
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    VersionName string
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    batchSize String
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    dataFormat String
    The format of the input data files.
    inputPaths List<String>
    The Cloud Storage location of the input data files. May contain wildcards.
    maxWorkerCount String
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    modelName String
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    outputDataFormat String
    Optional. Format of the output data files, defaults to JSON.
    outputPath String
    The output Google Cloud Storage location.
    region String
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    runtimeVersion String
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    signatureName String
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    uri String
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    versionName String
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    batchSize string
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    dataFormat string
    The format of the input data files.
    inputPaths string[]
    The Cloud Storage location of the input data files. May contain wildcards.
    maxWorkerCount string
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    modelName string
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    outputDataFormat string
    Optional. Format of the output data files, defaults to JSON.
    outputPath string
    The output Google Cloud Storage location.
    region string
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    runtimeVersion string
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    signatureName string
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    uri string
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    versionName string
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    batch_size str
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    data_format str
    The format of the input data files.
    input_paths Sequence[str]
    The Cloud Storage location of the input data files. May contain wildcards.
    max_worker_count str
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    model_name str
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    output_data_format str
    Optional. Format of the output data files, defaults to JSON.
    output_path str
    The output Google Cloud Storage location.
    region str
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    runtime_version str
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    signature_name str
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    uri str
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    version_name str
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"
    batchSize String
    Optional. Number of records per batch, defaults to 64. The service will buffer batch_size number of records in memory before invoking one Tensorflow prediction call internally. So take the record size and memory available into consideration when setting this parameter.
    dataFormat String
    The format of the input data files.
    inputPaths List<String>
    The Cloud Storage location of the input data files. May contain wildcards.
    maxWorkerCount String
    Optional. The maximum number of workers to be used for parallel processing. Defaults to 10 if not specified.
    modelName String
    Use this field if you want to use the default version for the specified model. The string must use the following format: "projects/YOUR_PROJECT/models/YOUR_MODEL"
    outputDataFormat String
    Optional. Format of the output data files, defaults to JSON.
    outputPath String
    The output Google Cloud Storage location.
    region String
    The Google Compute Engine region to run the prediction job in. See the available regions for AI Platform services.
    runtimeVersion String
    Optional. The AI Platform runtime version to use for this batch prediction. If not set, AI Platform will pick the runtime version used during the CreateVersion request for this model version, or choose the latest stable version when model version information is not available such as when the model is specified by uri.
    signatureName String
    Optional. The name of the signature defined in the SavedModel to use for this job. Please refer to SavedModel for information about how to use signatures. Defaults to DEFAULT_SERVING_SIGNATURE_DEF_KEY , which is "serving_default".
    uri String
    Use this field if you want to specify a Google Cloud Storage path for the model to use.
    versionName String
    Use this field if you want to specify a version of the model to use. The string is formatted the same way as model_version, with the addition of the version information: "projects/YOUR_PROJECT/models/YOUR_MODEL/versions/YOUR_VERSION"

    GoogleCloudMlV1__PredictionOutput, GoogleCloudMlV1__PredictionOutputArgs

    ErrorCount string
    The number of data instances which resulted in errors.
    NodeHours double
    Node hours used by the batch prediction job.
    OutputPath string
    The output Google Cloud Storage location provided at the job creation time.
    PredictionCount string
    The number of generated predictions.
    ErrorCount string
    The number of data instances which resulted in errors.
    NodeHours float64
    Node hours used by the batch prediction job.
    OutputPath string
    The output Google Cloud Storage location provided at the job creation time.
    PredictionCount string
    The number of generated predictions.
    errorCount String
    The number of data instances which resulted in errors.
    nodeHours Double
    Node hours used by the batch prediction job.
    outputPath String
    The output Google Cloud Storage location provided at the job creation time.
    predictionCount String
    The number of generated predictions.
    errorCount string
    The number of data instances which resulted in errors.
    nodeHours number
    Node hours used by the batch prediction job.
    outputPath string
    The output Google Cloud Storage location provided at the job creation time.
    predictionCount string
    The number of generated predictions.
    error_count str
    The number of data instances which resulted in errors.
    node_hours float
    Node hours used by the batch prediction job.
    output_path str
    The output Google Cloud Storage location provided at the job creation time.
    prediction_count str
    The number of generated predictions.
    errorCount String
    The number of data instances which resulted in errors.
    nodeHours Number
    Node hours used by the batch prediction job.
    outputPath String
    The output Google Cloud Storage location provided at the job creation time.
    predictionCount String
    The number of generated predictions.

    GoogleCloudMlV1__PredictionOutputResponse, GoogleCloudMlV1__PredictionOutputResponseArgs

    ErrorCount string
    The number of data instances which resulted in errors.
    NodeHours double
    Node hours used by the batch prediction job.
    OutputPath string
    The output Google Cloud Storage location provided at the job creation time.
    PredictionCount string
    The number of generated predictions.
    ErrorCount string
    The number of data instances which resulted in errors.
    NodeHours float64
    Node hours used by the batch prediction job.
    OutputPath string
    The output Google Cloud Storage location provided at the job creation time.
    PredictionCount string
    The number of generated predictions.
    errorCount String
    The number of data instances which resulted in errors.
    nodeHours Double
    Node hours used by the batch prediction job.
    outputPath String
    The output Google Cloud Storage location provided at the job creation time.
    predictionCount String
    The number of generated predictions.
    errorCount string
    The number of data instances which resulted in errors.
    nodeHours number
    Node hours used by the batch prediction job.
    outputPath string
    The output Google Cloud Storage location provided at the job creation time.
    predictionCount string
    The number of generated predictions.
    error_count str
    The number of data instances which resulted in errors.
    node_hours float
    Node hours used by the batch prediction job.
    output_path str
    The output Google Cloud Storage location provided at the job creation time.
    prediction_count str
    The number of generated predictions.
    errorCount String
    The number of data instances which resulted in errors.
    nodeHours Number
    Node hours used by the batch prediction job.
    outputPath String
    The output Google Cloud Storage location provided at the job creation time.
    predictionCount String
    The number of generated predictions.

    GoogleCloudMlV1__ReplicaConfig, GoogleCloudMlV1__ReplicaConfigArgs

    AcceleratorConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__AcceleratorConfig
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    ContainerArgs List<string>
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    ContainerCommand List<string>
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    DiskConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__DiskConfig
    Represents the configuration of disk options.
    ImageUri string
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    TpuTfVersion string
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    AcceleratorConfig GoogleCloudMlV1__AcceleratorConfig
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    ContainerArgs []string
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    ContainerCommand []string
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    DiskConfig GoogleCloudMlV1__DiskConfig
    Represents the configuration of disk options.
    ImageUri string
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    TpuTfVersion string
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    acceleratorConfig GoogleCloudMlV1__AcceleratorConfig
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    containerArgs List<String>
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    containerCommand List<String>
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    diskConfig GoogleCloudMlV1__DiskConfig
    Represents the configuration of disk options.
    imageUri String
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    tpuTfVersion String
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    acceleratorConfig GoogleCloudMlV1__AcceleratorConfig
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    containerArgs string[]
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    containerCommand string[]
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    diskConfig GoogleCloudMlV1__DiskConfig
    Represents the configuration of disk options.
    imageUri string
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    tpuTfVersion string
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    accelerator_config GoogleCloudMlV1AcceleratorConfig
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    container_args Sequence[str]
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    container_command Sequence[str]
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    disk_config GoogleCloudMlV1DiskConfig
    Represents the configuration of disk options.
    image_uri str
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    tpu_tf_version str
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    acceleratorConfig Property Map
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    containerArgs List<String>
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    containerCommand List<String>
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    diskConfig Property Map
    Represents the configuration of disk options.
    imageUri String
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    tpuTfVersion String
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.

    GoogleCloudMlV1__ReplicaConfigResponse, GoogleCloudMlV1__ReplicaConfigResponseArgs

    AcceleratorConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__AcceleratorConfigResponse
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    ContainerArgs List<string>
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    ContainerCommand List<string>
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    DiskConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__DiskConfigResponse
    Represents the configuration of disk options.
    ImageUri string
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    TpuTfVersion string
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    AcceleratorConfig GoogleCloudMlV1__AcceleratorConfigResponse
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    ContainerArgs []string
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    ContainerCommand []string
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    DiskConfig GoogleCloudMlV1__DiskConfigResponse
    Represents the configuration of disk options.
    ImageUri string
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    TpuTfVersion string
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    acceleratorConfig GoogleCloudMlV1__AcceleratorConfigResponse
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    containerArgs List<String>
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    containerCommand List<String>
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    diskConfig GoogleCloudMlV1__DiskConfigResponse
    Represents the configuration of disk options.
    imageUri String
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    tpuTfVersion String
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    acceleratorConfig GoogleCloudMlV1__AcceleratorConfigResponse
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    containerArgs string[]
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    containerCommand string[]
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    diskConfig GoogleCloudMlV1__DiskConfigResponse
    Represents the configuration of disk options.
    imageUri string
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    tpuTfVersion string
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    accelerator_config GoogleCloudMlV1AcceleratorConfigResponse
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    container_args Sequence[str]
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    container_command Sequence[str]
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    disk_config GoogleCloudMlV1DiskConfigResponse
    Represents the configuration of disk options.
    image_uri str
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    tpu_tf_version str
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.
    acceleratorConfig Property Map
    Represents the type and number of accelerators used by the replica. Learn about restrictions on accelerator configurations for training.
    containerArgs List<String>
    Arguments to the entrypoint command. The following rules apply for container_command and container_args: - If you do not supply command or args: The defaults defined in the Docker image are used. - If you supply a command but no args: The default EntryPoint and the default Cmd defined in the Docker image are ignored. Your command is run without any arguments. - If you supply only args: The default Entrypoint defined in the Docker image is run with the args that you supplied. - If you supply a command and args: The default Entrypoint and the default Cmd defined in the Docker image are ignored. Your command is run with your args. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    containerCommand List<String>
    The command with which the replica's custom container is run. If provided, it will override default ENTRYPOINT of the docker image. If not provided, the docker image's ENTRYPOINT is used. It cannot be set if custom container image is not provided. Note that this field and [TrainingInput.args] are mutually exclusive, i.e., both cannot be set at the same time.
    diskConfig Property Map
    Represents the configuration of disk options.
    imageUri String
    The Docker image to run on the replica. This image must be in Container Registry. Learn more about configuring custom containers.
    tpuTfVersion String
    The AI Platform runtime version that includes a TensorFlow version matching the one used in the custom container. This field is required if the replica is a TPU worker that uses a custom container. Otherwise, do not specify this field. This must be a runtime version that currently supports training with TPUs. Note that the version of TensorFlow included in a runtime version may differ from the numbering of the runtime version itself, because it may have a different patch version. In this field, you must specify the runtime version (TensorFlow minor version). For example, if your custom container runs TensorFlow 1.x.y, specify 1.x.

    GoogleCloudMlV1__Scheduling, GoogleCloudMlV1__SchedulingArgs

    MaxRunningTime string
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    MaxWaitTime string
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    Priority int
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    MaxRunningTime string
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    MaxWaitTime string
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    Priority int
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    maxRunningTime String
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    maxWaitTime String
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    priority Integer
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    maxRunningTime string
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    maxWaitTime string
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    priority number
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    max_running_time str
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    max_wait_time str
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    priority int
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    maxRunningTime String
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    maxWaitTime String
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    priority Number
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.

    GoogleCloudMlV1__SchedulingResponse, GoogleCloudMlV1__SchedulingResponseArgs

    MaxRunningTime string
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    MaxWaitTime string
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    Priority int
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    MaxRunningTime string
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    MaxWaitTime string
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    Priority int
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    maxRunningTime String
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    maxWaitTime String
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    priority Integer
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    maxRunningTime string
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    maxWaitTime string
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    priority number
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    max_running_time str
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    max_wait_time str
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    priority int
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.
    maxRunningTime String
    Optional. The maximum job running time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, this field defaults to 604800s (seven days). If the training job is still running after this duration, AI Platform Training cancels it. The duration is measured from when the job enters the RUNNING state; therefore it does not overlap with the duration limited by Scheduling.max_wait_time. For example, if you want to ensure your job runs for no more than 2 hours, set this field to 7200s (2 hours * 60 minutes / hour * 60 seconds / minute). If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxRunningTime: 7200s
    maxWaitTime String
    Optional. The maximum job wait time, expressed in seconds. The field can contain up to nine fractional digits, terminated by s. If not specified, there is no limit to the wait time. The minimum for this field is 1800s (30 minutes). If the training job has not entered the RUNNING state after this duration, AI Platform Training cancels it. After the job begins running, it can no longer be cancelled due to the maximum wait time. Therefore the duration limited by this field does not overlap with the duration limited by Scheduling.max_running_time. For example, if the job temporarily stops running and retries due to a VM restart, this cannot lead to a maximum wait time cancellation. However, independently of this constraint, AI Platform Training might stop a job if there are too many retries due to exhausted resources in a region. The following example describes how you might use this field: To cancel your job if it doesn't start running within 1 hour, set this field to 3600s (1 hour * 60 minutes / hour * 60 seconds / minute). If the job is still in the QUEUED or PREPARING state after an hour of waiting, AI Platform Training cancels the job. If you submit your training job using the gcloud tool, you can specify this field in a config.yaml file. For example: yaml trainingInput: scheduling: maxWaitTime: 3600s
    priority Number
    Optional. Job scheduling will be based on this priority, which in the range [0, 1000]. The bigger the number, the higher the priority. Default to 0 if not set. If there are multiple jobs requesting same type of accelerators, the high priority job will be scheduled prior to ones with low priority.

    GoogleCloudMlV1__TrainingInput, GoogleCloudMlV1__TrainingInputArgs

    PackageUris List<string>
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    PythonModule string
    The Python module name to run after installing the packages.
    Region string
    The region to run the training job in. See the available regions for AI Platform Training.
    ScaleTier Pulumi.GoogleNative.Ml.V1.GoogleCloudMlV1__TrainingInputScaleTier
    Specifies the machine types, the number of replicas for workers and parameter servers.
    Args List<string>
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    EnableWebAccess bool
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    EncryptionConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__EncryptionConfig
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    EvaluatorConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    EvaluatorCount string
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    EvaluatorType string
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    Hyperparameters Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__HyperparameterSpec
    Optional. The set of Hyperparameters to tune.
    JobDir string
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    MasterConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    MasterType string
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    Network string
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    ParameterServerConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    ParameterServerCount string
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    ParameterServerType string
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    PythonVersion string
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    RuntimeVersion string
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    Scheduling Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__Scheduling
    Optional. Scheduling options for a training job.
    ServiceAccount string
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    UseChiefInTfConfig bool
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    WorkerConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    WorkerCount string
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    WorkerType string
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    PackageUris []string
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    PythonModule string
    The Python module name to run after installing the packages.
    Region string
    The region to run the training job in. See the available regions for AI Platform Training.
    ScaleTier GoogleCloudMlV1__TrainingInputScaleTier
    Specifies the machine types, the number of replicas for workers and parameter servers.
    Args []string
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    EnableWebAccess bool
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    EncryptionConfig GoogleCloudMlV1__EncryptionConfig
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    EvaluatorConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    EvaluatorCount string
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    EvaluatorType string
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    Hyperparameters GoogleCloudMlV1__HyperparameterSpec
    Optional. The set of Hyperparameters to tune.
    JobDir string
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    MasterConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    MasterType string
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    Network string
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    ParameterServerConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    ParameterServerCount string
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    ParameterServerType string
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    PythonVersion string
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    RuntimeVersion string
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    Scheduling GoogleCloudMlV1__Scheduling
    Optional. Scheduling options for a training job.
    ServiceAccount string
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    UseChiefInTfConfig bool
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    WorkerConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    WorkerCount string
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    WorkerType string
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    packageUris List<String>
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    pythonModule String
    The Python module name to run after installing the packages.
    region String
    The region to run the training job in. See the available regions for AI Platform Training.
    scaleTier GoogleCloudMlV1__TrainingInputScaleTier
    Specifies the machine types, the number of replicas for workers and parameter servers.
    args List<String>
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    enableWebAccess Boolean
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    encryptionConfig GoogleCloudMlV1__EncryptionConfig
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    evaluatorConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    evaluatorCount String
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    evaluatorType String
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    hyperparameters GoogleCloudMlV1__HyperparameterSpec
    Optional. The set of Hyperparameters to tune.
    jobDir String
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    masterConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    masterType String
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    network String
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    parameterServerConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    parameterServerCount String
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    parameterServerType String
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    pythonVersion String
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    runtimeVersion String
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    scheduling GoogleCloudMlV1__Scheduling
    Optional. Scheduling options for a training job.
    serviceAccount String
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    useChiefInTfConfig Boolean
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    workerConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    workerCount String
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    workerType String
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    packageUris string[]
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    pythonModule string
    The Python module name to run after installing the packages.
    region string
    The region to run the training job in. See the available regions for AI Platform Training.
    scaleTier GoogleCloudMlV1__TrainingInputScaleTier
    Specifies the machine types, the number of replicas for workers and parameter servers.
    args string[]
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    enableWebAccess boolean
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    encryptionConfig GoogleCloudMlV1__EncryptionConfig
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    evaluatorConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    evaluatorCount string
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    evaluatorType string
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    hyperparameters GoogleCloudMlV1__HyperparameterSpec
    Optional. The set of Hyperparameters to tune.
    jobDir string
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    masterConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    masterType string
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    network string
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    parameterServerConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    parameterServerCount string
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    parameterServerType string
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    pythonVersion string
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    runtimeVersion string
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    scheduling GoogleCloudMlV1__Scheduling
    Optional. Scheduling options for a training job.
    serviceAccount string
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    useChiefInTfConfig boolean
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    workerConfig GoogleCloudMlV1__ReplicaConfig
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    workerCount string
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    workerType string
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    package_uris Sequence[str]
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    python_module str
    The Python module name to run after installing the packages.
    region str
    The region to run the training job in. See the available regions for AI Platform Training.
    scale_tier GoogleCloudMlV1TrainingInputScaleTier
    Specifies the machine types, the number of replicas for workers and parameter servers.
    args Sequence[str]
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    enable_web_access bool
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    encryption_config GoogleCloudMlV1EncryptionConfig
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    evaluator_config GoogleCloudMlV1ReplicaConfig
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    evaluator_count str
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    evaluator_type str
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    hyperparameters GoogleCloudMlV1HyperparameterSpec
    Optional. The set of Hyperparameters to tune.
    job_dir str
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    master_config GoogleCloudMlV1ReplicaConfig
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    master_type str
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    network str
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    parameter_server_config GoogleCloudMlV1ReplicaConfig
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    parameter_server_count str
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    parameter_server_type str
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    python_version str
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    runtime_version str
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    scheduling GoogleCloudMlV1Scheduling
    Optional. Scheduling options for a training job.
    service_account str
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    use_chief_in_tf_config bool
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    worker_config GoogleCloudMlV1ReplicaConfig
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    worker_count str
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    worker_type str
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    packageUris List<String>
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    pythonModule String
    The Python module name to run after installing the packages.
    region String
    The region to run the training job in. See the available regions for AI Platform Training.
    scaleTier "BASIC" | "STANDARD_1" | "PREMIUM_1" | "BASIC_GPU" | "BASIC_TPU" | "CUSTOM"
    Specifies the machine types, the number of replicas for workers and parameter servers.
    args List<String>
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    enableWebAccess Boolean
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    encryptionConfig Property Map
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    evaluatorConfig Property Map
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    evaluatorCount String
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    evaluatorType String
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    hyperparameters Property Map
    Optional. The set of Hyperparameters to tune.
    jobDir String
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    masterConfig Property Map
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    masterType String
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    network String
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    parameterServerConfig Property Map
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    parameterServerCount String
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    parameterServerType String
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    pythonVersion String
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    runtimeVersion String
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    scheduling Property Map
    Optional. Scheduling options for a training job.
    serviceAccount String
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    useChiefInTfConfig Boolean
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    workerConfig Property Map
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    workerCount String
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    workerType String
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.

    GoogleCloudMlV1__TrainingInputResponse, GoogleCloudMlV1__TrainingInputResponseArgs

    Args List<string>
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    EnableWebAccess bool
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    EncryptionConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__EncryptionConfigResponse
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    EvaluatorConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    EvaluatorCount string
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    EvaluatorType string
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    Hyperparameters Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__HyperparameterSpecResponse
    Optional. The set of Hyperparameters to tune.
    JobDir string
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    MasterConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    MasterType string
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    Network string
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    PackageUris List<string>
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    ParameterServerConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    ParameterServerCount string
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    ParameterServerType string
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    PythonModule string
    The Python module name to run after installing the packages.
    PythonVersion string
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    Region string
    The region to run the training job in. See the available regions for AI Platform Training.
    RuntimeVersion string
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    ScaleTier string
    Specifies the machine types, the number of replicas for workers and parameter servers.
    Scheduling Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__SchedulingResponse
    Optional. Scheduling options for a training job.
    ServiceAccount string
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    UseChiefInTfConfig bool
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    WorkerConfig Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    WorkerCount string
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    WorkerType string
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    Args []string
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    EnableWebAccess bool
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    EncryptionConfig GoogleCloudMlV1__EncryptionConfigResponse
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    EvaluatorConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    EvaluatorCount string
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    EvaluatorType string
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    Hyperparameters GoogleCloudMlV1__HyperparameterSpecResponse
    Optional. The set of Hyperparameters to tune.
    JobDir string
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    MasterConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    MasterType string
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    Network string
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    PackageUris []string
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    ParameterServerConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    ParameterServerCount string
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    ParameterServerType string
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    PythonModule string
    The Python module name to run after installing the packages.
    PythonVersion string
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    Region string
    The region to run the training job in. See the available regions for AI Platform Training.
    RuntimeVersion string
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    ScaleTier string
    Specifies the machine types, the number of replicas for workers and parameter servers.
    Scheduling GoogleCloudMlV1__SchedulingResponse
    Optional. Scheduling options for a training job.
    ServiceAccount string
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    UseChiefInTfConfig bool
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    WorkerConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    WorkerCount string
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    WorkerType string
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    args List<String>
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    enableWebAccess Boolean
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    encryptionConfig GoogleCloudMlV1__EncryptionConfigResponse
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    evaluatorConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    evaluatorCount String
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    evaluatorType String
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    hyperparameters GoogleCloudMlV1__HyperparameterSpecResponse
    Optional. The set of Hyperparameters to tune.
    jobDir String
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    masterConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    masterType String
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    network String
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    packageUris List<String>
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    parameterServerConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    parameterServerCount String
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    parameterServerType String
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    pythonModule String
    The Python module name to run after installing the packages.
    pythonVersion String
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    region String
    The region to run the training job in. See the available regions for AI Platform Training.
    runtimeVersion String
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    scaleTier String
    Specifies the machine types, the number of replicas for workers and parameter servers.
    scheduling GoogleCloudMlV1__SchedulingResponse
    Optional. Scheduling options for a training job.
    serviceAccount String
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    useChiefInTfConfig Boolean
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    workerConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    workerCount String
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    workerType String
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    args string[]
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    enableWebAccess boolean
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    encryptionConfig GoogleCloudMlV1__EncryptionConfigResponse
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    evaluatorConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    evaluatorCount string
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    evaluatorType string
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    hyperparameters GoogleCloudMlV1__HyperparameterSpecResponse
    Optional. The set of Hyperparameters to tune.
    jobDir string
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    masterConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    masterType string
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    network string
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    packageUris string[]
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    parameterServerConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    parameterServerCount string
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    parameterServerType string
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    pythonModule string
    The Python module name to run after installing the packages.
    pythonVersion string
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    region string
    The region to run the training job in. See the available regions for AI Platform Training.
    runtimeVersion string
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    scaleTier string
    Specifies the machine types, the number of replicas for workers and parameter servers.
    scheduling GoogleCloudMlV1__SchedulingResponse
    Optional. Scheduling options for a training job.
    serviceAccount string
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    useChiefInTfConfig boolean
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    workerConfig GoogleCloudMlV1__ReplicaConfigResponse
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    workerCount string
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    workerType string
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    args Sequence[str]
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    enable_web_access bool
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    encryption_config GoogleCloudMlV1EncryptionConfigResponse
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    evaluator_config GoogleCloudMlV1ReplicaConfigResponse
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    evaluator_count str
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    evaluator_type str
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    hyperparameters GoogleCloudMlV1HyperparameterSpecResponse
    Optional. The set of Hyperparameters to tune.
    job_dir str
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    master_config GoogleCloudMlV1ReplicaConfigResponse
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    master_type str
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    network str
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    package_uris Sequence[str]
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    parameter_server_config GoogleCloudMlV1ReplicaConfigResponse
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    parameter_server_count str
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    parameter_server_type str
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    python_module str
    The Python module name to run after installing the packages.
    python_version str
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    region str
    The region to run the training job in. See the available regions for AI Platform Training.
    runtime_version str
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    scale_tier str
    Specifies the machine types, the number of replicas for workers and parameter servers.
    scheduling GoogleCloudMlV1SchedulingResponse
    Optional. Scheduling options for a training job.
    service_account str
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    use_chief_in_tf_config bool
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    worker_config GoogleCloudMlV1ReplicaConfigResponse
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    worker_count str
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    worker_type str
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.
    args List<String>
    Optional. Command-line arguments passed to the training application when it starts. If your job uses a custom container, then the arguments are passed to the container's ENTRYPOINT command.
    enableWebAccess Boolean
    Optional. Whether you want AI Platform Training to enable interactive shell access to training containers. If set to true, you can access interactive shells at the URIs given by TrainingOutput.web_access_uris or HyperparameterOutput.web_access_uris (within TrainingOutput.trials).
    encryptionConfig Property Map
    Optional. Options for using customer-managed encryption keys (CMEK) to protect resources created by a training job, instead of using Google's default encryption. If this is set, then all resources created by the training job will be encrypted with the customer-managed encryption key that you specify. Learn how and when to use CMEK with AI Platform Training.
    evaluatorConfig Property Map
    Optional. The configuration for evaluators. You should only set evaluatorConfig.acceleratorConfig if evaluatorType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set evaluatorConfig.imageUri only if you build a custom image for your evaluator. If evaluatorConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    evaluatorCount String
    Optional. The number of evaluator replicas to use for the training job. Each replica in the cluster will be of the type specified in evaluator_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set evaluator_type. The default value is zero.
    evaluatorType String
    Optional. Specifies the type of virtual machine to use for your training job's evaluator nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and evaluatorCount is greater than zero.
    hyperparameters Property Map
    Optional. The set of Hyperparameters to tune.
    jobDir String
    Optional. A Google Cloud Storage path in which to store training outputs and other data needed for training. This path is passed to your TensorFlow program as the '--job-dir' command-line argument. The benefit of specifying this field is that Cloud ML validates the path for use in training.
    masterConfig Property Map
    Optional. The configuration for your master worker. You should only set masterConfig.acceleratorConfig if masterType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set masterConfig.imageUri only if you build a custom image. Only one of masterConfig.imageUri and runtimeVersion should be set. Learn more about configuring custom containers.
    masterType String
    Optional. Specifies the type of virtual machine to use for your training job's master worker. You must specify this field when scaleTier is set to CUSTOM. You can use certain Compute Engine machine types directly in this field. See the list of compatible Compute Engine machine types. Alternatively, you can use the certain legacy machine types in this field. See the list of legacy machine types. Finally, if you want to use a TPU for training, specify cloud_tpu in this field. Learn more about the special configuration options for training with TPUs.
    network String
    Optional. The full name of the Compute Engine network to which the Job is peered. For example, projects/12345/global/networks/myVPC. The format of this field is projects/{project}/global/networks/{network}, where {project} is a project number (like 12345) and {network} is network name. Private services access must already be configured for the network. If left unspecified, the Job is not peered with any network. Learn about using VPC Network Peering..
    packageUris List<String>
    The Google Cloud Storage location of the packages with the training program and any additional dependencies. The maximum number of package URIs is 100.
    parameterServerConfig Property Map
    Optional. The configuration for parameter servers. You should only set parameterServerConfig.acceleratorConfig if parameterServerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set parameterServerConfig.imageUri only if you build a custom image for your parameter server. If parameterServerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    parameterServerCount String
    Optional. The number of parameter server replicas to use for the training job. Each replica in the cluster will be of the type specified in parameter_server_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set parameter_server_type. The default value is zero.
    parameterServerType String
    Optional. Specifies the type of virtual machine to use for your training job's parameter server. The supported values are the same as those described in the entry for master_type. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. This value must be present when scaleTier is set to CUSTOM and parameter_server_count is greater than zero.
    pythonModule String
    The Python module name to run after installing the packages.
    pythonVersion String
    Optional. The version of Python used in training. You must either specify this field or specify masterConfig.imageUri. The following Python versions are available: * Python '3.7' is available when runtime_version is set to '1.15' or later. * Python '3.5' is available when runtime_version is set to a version from '1.4' to '1.14'. * Python '2.7' is available when runtime_version is set to '1.15' or earlier. Read more about the Python versions available for each runtime version.
    region String
    The region to run the training job in. See the available regions for AI Platform Training.
    runtimeVersion String
    Optional. The AI Platform runtime version to use for training. You must either specify this field or specify masterConfig.imageUri. For more information, see the runtime version list and learn how to manage runtime versions.
    scaleTier String
    Specifies the machine types, the number of replicas for workers and parameter servers.
    scheduling Property Map
    Optional. Scheduling options for a training job.
    serviceAccount String
    Optional. The email address of a service account to use when running the training appplication. You must have the iam.serviceAccounts.actAs permission for the specified service account. In addition, the AI Platform Training Google-managed service account must have the roles/iam.serviceAccountAdmin role for the specified service account. Learn more about configuring a service account. If not specified, the AI Platform Training Google-managed service account is used by default.
    useChiefInTfConfig Boolean
    Optional. Use chief instead of master in the TF_CONFIG environment variable when training with a custom container. Defaults to false. Learn more about this field. This field has no effect for training jobs that don't use a custom container.
    workerConfig Property Map
    Optional. The configuration for workers. You should only set workerConfig.acceleratorConfig if workerType is set to a Compute Engine machine type. Learn about restrictions on accelerator configurations for training. Set workerConfig.imageUri only if you build a custom image for your worker. If workerConfig.imageUri has not been set, AI Platform uses the value of masterConfig.imageUri. Learn more about configuring custom containers.
    workerCount String
    Optional. The number of worker replicas to use for the training job. Each replica in the cluster will be of the type specified in worker_type. This value can only be used when scale_tier is set to CUSTOM. If you set this value, you must also set worker_type. The default value is zero.
    workerType String
    Optional. Specifies the type of virtual machine to use for your training job's worker nodes. The supported values are the same as those described in the entry for masterType. This value must be consistent with the category of machine type that masterType uses. In other words, both must be Compute Engine machine types or both must be legacy machine types. If you use cloud_tpu for this value, see special instructions for configuring a custom TPU machine. This value must be present when scaleTier is set to CUSTOM and workerCount is greater than zero.

    GoogleCloudMlV1__TrainingInputScaleTier, GoogleCloudMlV1__TrainingInputScaleTierArgs

    Basic
    BASICA single worker instance. This tier is suitable for learning how to use Cloud ML, and for experimenting with new models using small datasets.
    Standard1
    STANDARD_1Many workers and a few parameter servers.
    Premium1
    PREMIUM_1A large number of workers with many parameter servers.
    BasicGpu
    BASIC_GPUA single worker instance with a GPU.
    BasicTpu
    BASIC_TPUA single worker instance with a Cloud TPU.
    Custom
    CUSTOMThe CUSTOM tier is not a set tier, but rather enables you to use your own cluster specification. When you use this tier, set values to configure your processing cluster according to these guidelines: * You must set TrainingInput.masterType to specify the type of machine to use for your master node. This is the only required setting. * You may set TrainingInput.workerCount to specify the number of workers to use. If you specify one or more workers, you must also set TrainingInput.workerType to specify the type of machine to use for your worker nodes. * You may set TrainingInput.parameterServerCount to specify the number of parameter servers to use. If you specify one or more parameter servers, you must also set TrainingInput.parameterServerType to specify the type of machine to use for your parameter servers. Note that all of your workers must use the same machine type, which can be different from your parameter server type and master type. Your parameter servers must likewise use the same machine type, which can be different from your worker type and master type.
    GoogleCloudMlV1__TrainingInputScaleTierBasic
    BASICA single worker instance. This tier is suitable for learning how to use Cloud ML, and for experimenting with new models using small datasets.
    GoogleCloudMlV1__TrainingInputScaleTierStandard1
    STANDARD_1Many workers and a few parameter servers.
    GoogleCloudMlV1__TrainingInputScaleTierPremium1
    PREMIUM_1A large number of workers with many parameter servers.
    GoogleCloudMlV1__TrainingInputScaleTierBasicGpu
    BASIC_GPUA single worker instance with a GPU.
    GoogleCloudMlV1__TrainingInputScaleTierBasicTpu
    BASIC_TPUA single worker instance with a Cloud TPU.
    GoogleCloudMlV1__TrainingInputScaleTierCustom
    CUSTOMThe CUSTOM tier is not a set tier, but rather enables you to use your own cluster specification. When you use this tier, set values to configure your processing cluster according to these guidelines: * You must set TrainingInput.masterType to specify the type of machine to use for your master node. This is the only required setting. * You may set TrainingInput.workerCount to specify the number of workers to use. If you specify one or more workers, you must also set TrainingInput.workerType to specify the type of machine to use for your worker nodes. * You may set TrainingInput.parameterServerCount to specify the number of parameter servers to use. If you specify one or more parameter servers, you must also set TrainingInput.parameterServerType to specify the type of machine to use for your parameter servers. Note that all of your workers must use the same machine type, which can be different from your parameter server type and master type. Your parameter servers must likewise use the same machine type, which can be different from your worker type and master type.
    Basic
    BASICA single worker instance. This tier is suitable for learning how to use Cloud ML, and for experimenting with new models using small datasets.
    Standard1
    STANDARD_1Many workers and a few parameter servers.
    Premium1
    PREMIUM_1A large number of workers with many parameter servers.
    BasicGpu
    BASIC_GPUA single worker instance with a GPU.
    BasicTpu
    BASIC_TPUA single worker instance with a Cloud TPU.
    Custom
    CUSTOMThe CUSTOM tier is not a set tier, but rather enables you to use your own cluster specification. When you use this tier, set values to configure your processing cluster according to these guidelines: * You must set TrainingInput.masterType to specify the type of machine to use for your master node. This is the only required setting. * You may set TrainingInput.workerCount to specify the number of workers to use. If you specify one or more workers, you must also set TrainingInput.workerType to specify the type of machine to use for your worker nodes. * You may set TrainingInput.parameterServerCount to specify the number of parameter servers to use. If you specify one or more parameter servers, you must also set TrainingInput.parameterServerType to specify the type of machine to use for your parameter servers. Note that all of your workers must use the same machine type, which can be different from your parameter server type and master type. Your parameter servers must likewise use the same machine type, which can be different from your worker type and master type.
    Basic
    BASICA single worker instance. This tier is suitable for learning how to use Cloud ML, and for experimenting with new models using small datasets.
    Standard1
    STANDARD_1Many workers and a few parameter servers.
    Premium1
    PREMIUM_1A large number of workers with many parameter servers.
    BasicGpu
    BASIC_GPUA single worker instance with a GPU.
    BasicTpu
    BASIC_TPUA single worker instance with a Cloud TPU.
    Custom
    CUSTOMThe CUSTOM tier is not a set tier, but rather enables you to use your own cluster specification. When you use this tier, set values to configure your processing cluster according to these guidelines: * You must set TrainingInput.masterType to specify the type of machine to use for your master node. This is the only required setting. * You may set TrainingInput.workerCount to specify the number of workers to use. If you specify one or more workers, you must also set TrainingInput.workerType to specify the type of machine to use for your worker nodes. * You may set TrainingInput.parameterServerCount to specify the number of parameter servers to use. If you specify one or more parameter servers, you must also set TrainingInput.parameterServerType to specify the type of machine to use for your parameter servers. Note that all of your workers must use the same machine type, which can be different from your parameter server type and master type. Your parameter servers must likewise use the same machine type, which can be different from your worker type and master type.
    BASIC
    BASICA single worker instance. This tier is suitable for learning how to use Cloud ML, and for experimenting with new models using small datasets.
    STANDARD1
    STANDARD_1Many workers and a few parameter servers.
    PREMIUM1
    PREMIUM_1A large number of workers with many parameter servers.
    BASIC_GPU
    BASIC_GPUA single worker instance with a GPU.
    BASIC_TPU
    BASIC_TPUA single worker instance with a Cloud TPU.
    CUSTOM
    CUSTOMThe CUSTOM tier is not a set tier, but rather enables you to use your own cluster specification. When you use this tier, set values to configure your processing cluster according to these guidelines: * You must set TrainingInput.masterType to specify the type of machine to use for your master node. This is the only required setting. * You may set TrainingInput.workerCount to specify the number of workers to use. If you specify one or more workers, you must also set TrainingInput.workerType to specify the type of machine to use for your worker nodes. * You may set TrainingInput.parameterServerCount to specify the number of parameter servers to use. If you specify one or more parameter servers, you must also set TrainingInput.parameterServerType to specify the type of machine to use for your parameter servers. Note that all of your workers must use the same machine type, which can be different from your parameter server type and master type. Your parameter servers must likewise use the same machine type, which can be different from your worker type and master type.
    "BASIC"
    BASICA single worker instance. This tier is suitable for learning how to use Cloud ML, and for experimenting with new models using small datasets.
    "STANDARD_1"
    STANDARD_1Many workers and a few parameter servers.
    "PREMIUM_1"
    PREMIUM_1A large number of workers with many parameter servers.
    "BASIC_GPU"
    BASIC_GPUA single worker instance with a GPU.
    "BASIC_TPU"
    BASIC_TPUA single worker instance with a Cloud TPU.
    "CUSTOM"
    CUSTOMThe CUSTOM tier is not a set tier, but rather enables you to use your own cluster specification. When you use this tier, set values to configure your processing cluster according to these guidelines: * You must set TrainingInput.masterType to specify the type of machine to use for your master node. This is the only required setting. * You may set TrainingInput.workerCount to specify the number of workers to use. If you specify one or more workers, you must also set TrainingInput.workerType to specify the type of machine to use for your worker nodes. * You may set TrainingInput.parameterServerCount to specify the number of parameter servers to use. If you specify one or more parameter servers, you must also set TrainingInput.parameterServerType to specify the type of machine to use for your parameter servers. Note that all of your workers must use the same machine type, which can be different from your parameter server type and master type. Your parameter servers must likewise use the same machine type, which can be different from your worker type and master type.

    GoogleCloudMlV1__TrainingOutput, GoogleCloudMlV1__TrainingOutputArgs

    BuiltInAlgorithmOutput Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    CompletedTrialCount string
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    ConsumedMLUnits double
    The amount of ML units consumed by the job.
    HyperparameterMetricTag string
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    IsBuiltInAlgorithmJob bool
    Whether this job is a built-in Algorithm job.
    IsHyperparameterTuningJob bool
    Whether this job is a hyperparameter tuning job.
    Trials List<Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__HyperparameterOutput>
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    BuiltInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    CompletedTrialCount string
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    ConsumedMLUnits float64
    The amount of ML units consumed by the job.
    HyperparameterMetricTag string
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    IsBuiltInAlgorithmJob bool
    Whether this job is a built-in Algorithm job.
    IsHyperparameterTuningJob bool
    Whether this job is a hyperparameter tuning job.
    Trials []GoogleCloudMlV1__HyperparameterOutput
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    builtInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    completedTrialCount String
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    consumedMLUnits Double
    The amount of ML units consumed by the job.
    hyperparameterMetricTag String
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    isBuiltInAlgorithmJob Boolean
    Whether this job is a built-in Algorithm job.
    isHyperparameterTuningJob Boolean
    Whether this job is a hyperparameter tuning job.
    trials List<GoogleCloudMlV1__HyperparameterOutput>
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    builtInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    completedTrialCount string
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    consumedMLUnits number
    The amount of ML units consumed by the job.
    hyperparameterMetricTag string
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    isBuiltInAlgorithmJob boolean
    Whether this job is a built-in Algorithm job.
    isHyperparameterTuningJob boolean
    Whether this job is a hyperparameter tuning job.
    trials GoogleCloudMlV1__HyperparameterOutput[]
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    built_in_algorithm_output GoogleCloudMlV1BuiltInAlgorithmOutput
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    completed_trial_count str
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    consumed_ml_units float
    The amount of ML units consumed by the job.
    hyperparameter_metric_tag str
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    is_built_in_algorithm_job bool
    Whether this job is a built-in Algorithm job.
    is_hyperparameter_tuning_job bool
    Whether this job is a hyperparameter tuning job.
    trials Sequence[GoogleCloudMlV1HyperparameterOutput]
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    builtInAlgorithmOutput Property Map
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    completedTrialCount String
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    consumedMLUnits Number
    The amount of ML units consumed by the job.
    hyperparameterMetricTag String
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    isBuiltInAlgorithmJob Boolean
    Whether this job is a built-in Algorithm job.
    isHyperparameterTuningJob Boolean
    Whether this job is a hyperparameter tuning job.
    trials List<Property Map>
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.

    GoogleCloudMlV1__TrainingOutputResponse, GoogleCloudMlV1__TrainingOutputResponseArgs

    BuiltInAlgorithmOutput Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    CompletedTrialCount string
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    ConsumedMLUnits double
    The amount of ML units consumed by the job.
    HyperparameterMetricTag string
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    IsBuiltInAlgorithmJob bool
    Whether this job is a built-in Algorithm job.
    IsHyperparameterTuningJob bool
    Whether this job is a hyperparameter tuning job.
    Trials List<Pulumi.GoogleNative.Ml.V1.Inputs.GoogleCloudMlV1__HyperparameterOutputResponse>
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    WebAccessUris Dictionary<string, string>
    URIs for accessing interactive shells (one URI for each training node). Only available if training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    BuiltInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    CompletedTrialCount string
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    ConsumedMLUnits float64
    The amount of ML units consumed by the job.
    HyperparameterMetricTag string
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    IsBuiltInAlgorithmJob bool
    Whether this job is a built-in Algorithm job.
    IsHyperparameterTuningJob bool
    Whether this job is a hyperparameter tuning job.
    Trials []GoogleCloudMlV1__HyperparameterOutputResponse
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    WebAccessUris map[string]string
    URIs for accessing interactive shells (one URI for each training node). Only available if training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    builtInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    completedTrialCount String
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    consumedMLUnits Double
    The amount of ML units consumed by the job.
    hyperparameterMetricTag String
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    isBuiltInAlgorithmJob Boolean
    Whether this job is a built-in Algorithm job.
    isHyperparameterTuningJob Boolean
    Whether this job is a hyperparameter tuning job.
    trials List<GoogleCloudMlV1__HyperparameterOutputResponse>
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    webAccessUris Map<String,String>
    URIs for accessing interactive shells (one URI for each training node). Only available if training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    builtInAlgorithmOutput GoogleCloudMlV1__BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    completedTrialCount string
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    consumedMLUnits number
    The amount of ML units consumed by the job.
    hyperparameterMetricTag string
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    isBuiltInAlgorithmJob boolean
    Whether this job is a built-in Algorithm job.
    isHyperparameterTuningJob boolean
    Whether this job is a hyperparameter tuning job.
    trials GoogleCloudMlV1__HyperparameterOutputResponse[]
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    webAccessUris {[key: string]: string}
    URIs for accessing interactive shells (one URI for each training node). Only available if training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    built_in_algorithm_output GoogleCloudMlV1BuiltInAlgorithmOutputResponse
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    completed_trial_count str
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    consumed_ml_units float
    The amount of ML units consumed by the job.
    hyperparameter_metric_tag str
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    is_built_in_algorithm_job bool
    Whether this job is a built-in Algorithm job.
    is_hyperparameter_tuning_job bool
    Whether this job is a hyperparameter tuning job.
    trials Sequence[GoogleCloudMlV1HyperparameterOutputResponse]
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    web_access_uris Mapping[str, str]
    URIs for accessing interactive shells (one URI for each training node). Only available if training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.
    builtInAlgorithmOutput Property Map
    Details related to built-in algorithms jobs. Only set for built-in algorithms jobs.
    completedTrialCount String
    The number of hyperparameter tuning trials that completed successfully. Only set for hyperparameter tuning jobs.
    consumedMLUnits Number
    The amount of ML units consumed by the job.
    hyperparameterMetricTag String
    The TensorFlow summary tag name used for optimizing hyperparameter tuning trials. See HyperparameterSpec.hyperparameterMetricTag for more information. Only set for hyperparameter tuning jobs.
    isBuiltInAlgorithmJob Boolean
    Whether this job is a built-in Algorithm job.
    isHyperparameterTuningJob Boolean
    Whether this job is a hyperparameter tuning job.
    trials List<Property Map>
    Results for individual Hyperparameter trials. Only set for hyperparameter tuning jobs.
    webAccessUris Map<String>
    URIs for accessing interactive shells (one URI for each training node). Only available if training_input.enable_web_access is true. The keys are names of each node in the training job; for example, master-replica-0 for the master node, worker-replica-0 for the first worker, and ps-replica-0 for the first parameter server. The values are the URIs for each node's interactive shell.

    Package Details

    Repository
    Google Cloud Native pulumi/pulumi-google-native
    License
    Apache-2.0
    google-native logo

    Google Cloud Native is in preview. Google Cloud Classic is fully supported.

    Google Cloud Native v0.32.0 published on Wednesday, Nov 29, 2023 by Pulumi
    pulumi/pulumi-google-native