tuning_jobs

Creates, updates, deletes, gets or lists a tuning_jobs resource.

Overview

Name	tuning_jobs
Type	Resource
Id	google.aiplatform.tuning_jobs

Fields

The following fields are returned by SELECT queries:

get

Name	Datatype	Description
name	string	Output only. Identifier. Resource name of a TuningJob. Format: projects/{project}/locations/{location}/tuningJobs/{tuning_job}
baseModel	string	The base model that is being tuned. See Supported models.
createTime	string (google-datetime)	Output only. Time when the TuningJob was created.
description	string	Optional. The description of the TuningJob.
encryptionSpec	object	Customer-managed encryption key options for a TuningJob. If this is set, then all resources created by the TuningJob will be encrypted with the provided encryption key. (id: GoogleCloudAiplatformV1EncryptionSpec)
endTime	string (google-datetime)	Output only. Time when the TuningJob entered any of the following JobStates: JOB_STATE_SUCCEEDED, JOB_STATE_FAILED, JOB_STATE_CANCELLED, JOB_STATE_EXPIRED.
error	object	The Status type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by gRPC. Each Status message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the API Design Guide. (id: GoogleRpcStatus)
evaluateDatasetRuns	array	Output only. Evaluation runs for the Tuning Job.
experiment	string	Output only. The Experiment associated with this TuningJob.
labels	object	Optional. The labels with user-defined metadata to organize TuningJob and generated resources such as Model and Endpoint. Label keys and values can be no longer than 64 characters (Unicode codepoints), can only contain lowercase letters, numeric characters, underscores and dashes. International characters are allowed. See https://goo.gl/xmQnxf for more information and examples of labels.
preTunedModel	object	The pre-tuned model for continuous tuning. (id: GoogleCloudAiplatformV1PreTunedModel)
preferenceOptimizationSpec	object	Tuning Spec for Preference Optimization. (id: GoogleCloudAiplatformV1PreferenceOptimizationSpec)
serviceAccount	string	The service account that the tuningJob workload runs as. If not specified, the Vertex AI Secure Fine-Tuned Service Agent in the project will be used. See https://cloud.google.com/iam/docs/service-agents#vertex-ai-secure-fine-tuning-service-agent Users starting the pipeline must have the iam.serviceAccounts.actAs permission on this service account.
startTime	string (google-datetime)	Output only. Time when the TuningJob for the first time entered the JOB_STATE_RUNNING state.
state	string	Output only. The detailed state of the job. (JOB_STATE_UNSPECIFIED, JOB_STATE_QUEUED, JOB_STATE_PENDING, JOB_STATE_RUNNING, JOB_STATE_SUCCEEDED, JOB_STATE_FAILED, JOB_STATE_CANCELLING, JOB_STATE_CANCELLED, JOB_STATE_PAUSED, JOB_STATE_EXPIRED, JOB_STATE_UPDATING, JOB_STATE_PARTIALLY_SUCCEEDED)
supervisedTuningSpec	object	Tuning Spec for Supervised Fine Tuning. (id: GoogleCloudAiplatformV1SupervisedTuningSpec)
tunedModel	object	Output only. The tuned model resources associated with this TuningJob. (id: GoogleCloudAiplatformV1TunedModel)
tunedModelDisplayName	string	Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. For continuous tuning, tuned_model_display_name will by default use the same display name as the pre-tuned model. If a new display name is provided, the tuning job will create a new model instead of a new version.
tuningDataStats	object	Output only. The tuning data statistics associated with this TuningJob. (id: GoogleCloudAiplatformV1TuningDataStats)
updateTime	string (google-datetime)	Output only. Time when the TuningJob was most recently updated.

list

Name	Datatype	Description
name	string	Output only. Identifier. Resource name of a TuningJob. Format: projects/{project}/locations/{location}/tuningJobs/{tuning_job}
baseModel	string	The base model that is being tuned. See Supported models.
createTime	string (google-datetime)	Output only. Time when the TuningJob was created.
description	string	Optional. The description of the TuningJob.
encryptionSpec	object	Customer-managed encryption key options for a TuningJob. If this is set, then all resources created by the TuningJob will be encrypted with the provided encryption key. (id: GoogleCloudAiplatformV1EncryptionSpec)
endTime	string (google-datetime)	Output only. Time when the TuningJob entered any of the following JobStates: JOB_STATE_SUCCEEDED, JOB_STATE_FAILED, JOB_STATE_CANCELLED, JOB_STATE_EXPIRED.
error	object	The Status type defines a logical error model that is suitable for different programming environments, including REST APIs and RPC APIs. It is used by gRPC. Each Status message contains three pieces of data: error code, error message, and error details. You can find out more about this error model and how to work with it in the API Design Guide. (id: GoogleRpcStatus)
evaluateDatasetRuns	array	Output only. Evaluation runs for the Tuning Job.
experiment	string	Output only. The Experiment associated with this TuningJob.
labels	object	Optional. The labels with user-defined metadata to organize TuningJob and generated resources such as Model and Endpoint. Label keys and values can be no longer than 64 characters (Unicode codepoints), can only contain lowercase letters, numeric characters, underscores and dashes. International characters are allowed. See https://goo.gl/xmQnxf for more information and examples of labels.
preTunedModel	object	The pre-tuned model for continuous tuning. (id: GoogleCloudAiplatformV1PreTunedModel)
preferenceOptimizationSpec	object	Tuning Spec for Preference Optimization. (id: GoogleCloudAiplatformV1PreferenceOptimizationSpec)
serviceAccount	string	The service account that the tuningJob workload runs as. If not specified, the Vertex AI Secure Fine-Tuned Service Agent in the project will be used. See https://cloud.google.com/iam/docs/service-agents#vertex-ai-secure-fine-tuning-service-agent Users starting the pipeline must have the iam.serviceAccounts.actAs permission on this service account.
startTime	string (google-datetime)	Output only. Time when the TuningJob for the first time entered the JOB_STATE_RUNNING state.
state	string	Output only. The detailed state of the job. (JOB_STATE_UNSPECIFIED, JOB_STATE_QUEUED, JOB_STATE_PENDING, JOB_STATE_RUNNING, JOB_STATE_SUCCEEDED, JOB_STATE_FAILED, JOB_STATE_CANCELLING, JOB_STATE_CANCELLED, JOB_STATE_PAUSED, JOB_STATE_EXPIRED, JOB_STATE_UPDATING, JOB_STATE_PARTIALLY_SUCCEEDED)
supervisedTuningSpec	object	Tuning Spec for Supervised Fine Tuning. (id: GoogleCloudAiplatformV1SupervisedTuningSpec)
tunedModel	object	Output only. The tuned model resources associated with this TuningJob. (id: GoogleCloudAiplatformV1TunedModel)
tunedModelDisplayName	string	Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. For continuous tuning, tuned_model_display_name will by default use the same display name as the pre-tuned model. If a new display name is provided, the tuning job will create a new model instead of a new version.
tuningDataStats	object	Output only. The tuning data statistics associated with this TuningJob. (id: GoogleCloudAiplatformV1TuningDataStats)
updateTime	string (google-datetime)	Output only. Time when the TuningJob was most recently updated.

Methods

The following methods are available for this resource:

Name	Accessible by	Required Params	Optional Params	Description
get	select	projectsId, locationsId, tuningJobsId		Gets a tuning job.
list	select	projectsId, locationsId	pageToken, filter, pageSize	Lists tuning jobs in a location.
create	insert	projectsId, locationsId		Creates a tuning job. A created tuning job will be subsequently executed to start the model tuning process.
rebase_tuned_model	exec	projectsId, locationsId		Rebase a tuned model. A rebase operation takes a model that was previously tuned on a base model version, and retunes it on a new base model version. The rebase operation creates a new tuning job and a new tuned model.
cancel	exec	projectsId, locationsId, tuningJobsId		Cancels a tuning job. Starts an asynchronous cancellation request. The server makes a best effort to cancel the job, but success is not guaranteed. Clients can use GenAiTuningService.GetTuningJob or other methods to check whether the cancellation succeeded or whether the job completed despite cancellation. On successful cancellation, the tuning job is not deleted. Instead, its state is set to CANCELLED, and error is set to a status with a google.rpc.Status.code of 1, corresponding to Code.CANCELLED.

Parameters

Parameters can be passed in the WHERE clause of a query. Check the Methods section to see which parameters are required or optional for each operation.

Name	Datatype	Description
locationsId	string
projectsId	string
tuningJobsId	string
filter	string
pageSize	integer (int32)
pageToken	string

SELECT examples

get

Gets a tuning job.

SELECT
name,
baseModel,
createTime,
description,
encryptionSpec,
endTime,
error,
evaluateDatasetRuns,
experiment,
labels,
preTunedModel,
preferenceOptimizationSpec,
serviceAccount,
startTime,
state,
supervisedTuningSpec,
tunedModel,
tunedModelDisplayName,
tuningDataStats,
updateTime
FROM google.aiplatform.tuning_jobs
WHERE projectsId = '{{ projectsId }}' -- required
AND locationsId = '{{ locationsId }}' -- required
AND tuningJobsId = '{{ tuningJobsId }}' -- required
;

list

Lists tuning jobs in a location.

SELECT
name,
baseModel,
createTime,
description,
encryptionSpec,
endTime,
error,
evaluateDatasetRuns,
experiment,
labels,
preTunedModel,
preferenceOptimizationSpec,
serviceAccount,
startTime,
state,
supervisedTuningSpec,
tunedModel,
tunedModelDisplayName,
tuningDataStats,
updateTime
FROM google.aiplatform.tuning_jobs
WHERE projectsId = '{{ projectsId }}' -- required
AND locationsId = '{{ locationsId }}' -- required
AND pageToken = '{{ pageToken }}'
AND filter = '{{ filter }}'
AND pageSize = '{{ pageSize }}'
;

INSERT examples

create

Creates a tuning job. A created tuning job will be subsequently executed to start the model tuning process.

INSERT INTO google.aiplatform.tuning_jobs (
data__tunedModelDisplayName,
data__description,
data__baseModel,
data__labels,
data__supervisedTuningSpec,
data__encryptionSpec,
data__preferenceOptimizationSpec,
data__preTunedModel,
data__serviceAccount,
projectsId,
locationsId
)
SELECT 
'{{ tunedModelDisplayName }}',
'{{ description }}',
'{{ baseModel }}',
'{{ labels }}',
'{{ supervisedTuningSpec }}',
'{{ encryptionSpec }}',
'{{ preferenceOptimizationSpec }}',
'{{ preTunedModel }}',
'{{ serviceAccount }}',
'{{ projectsId }}',
'{{ locationsId }}'
RETURNING
name,
baseModel,
createTime,
description,
encryptionSpec,
endTime,
error,
evaluateDatasetRuns,
experiment,
labels,
preTunedModel,
preferenceOptimizationSpec,
serviceAccount,
startTime,
state,
supervisedTuningSpec,
tunedModel,
tunedModelDisplayName,
tuningDataStats,
updateTime
;

Manifest

# Description fields are for documentation purposes
- name: tuning_jobs
props:
  - name: projectsId
    value: "{{ projectsId }}"
    description: Required parameter for the tuning_jobs resource.
  - name: locationsId
    value: "{{ locationsId }}"
    description: Required parameter for the tuning_jobs resource.
  - name: tunedModelDisplayName
    value: "{{ tunedModelDisplayName }}"
    description: |
      Optional. The display name of the TunedModel. The name can be up to 128 characters long and can consist of any UTF-8 characters. For continuous tuning, tuned_model_display_name will by default use the same display name as the pre-tuned model. If a new display name is provided, the tuning job will create a new model instead of a new version.
  - name: description
    value: "{{ description }}"
    description: |
      Optional. The description of the TuningJob.
  - name: baseModel
    value: "{{ baseModel }}"
    description: |
      The base model that is being tuned. See [Supported models](https://cloud.google.com/vertex-ai/generative-ai/docs/model-reference/tuning#supported_models).
  - name: labels
    value: "{{ labels }}"
    description: |
      Optional. The labels with user-defined metadata to organize TuningJob and generated resources such as Model and Endpoint. Label keys and values can be no longer than 64 characters (Unicode codepoints), can only contain lowercase letters, numeric characters, underscores and dashes. International characters are allowed. See https://goo.gl/xmQnxf for more information and examples of labels.
  - name: supervisedTuningSpec
    description: |
      Tuning Spec for Supervised Fine Tuning.
    value:
      exportLastCheckpointOnly: {{ exportLastCheckpointOnly }}
      validationDatasetUri: "{{ validationDatasetUri }}"
      trainingDatasetUri: "{{ trainingDatasetUri }}"
      evaluationConfig:
        metrics:
          - pointwiseMetricSpec:
              metricPromptTemplate: "{{ metricPromptTemplate }}"
              systemInstruction: "{{ systemInstruction }}"
              customOutputFormatConfig:
                returnRawOutput: {{ returnRawOutput }}
            bleuSpec:
              useEffectiveOrder: {{ useEffectiveOrder }}
            rougeSpec:
              useStemmer: {{ useStemmer }}
              rougeType: "{{ rougeType }}"
              splitSummaries: {{ splitSummaries }}
            exactMatchSpec: "{{ exactMatchSpec }}"
            computationBasedMetricSpec:
              type: "{{ type }}"
              parameters: "{{ parameters }}"
            predefinedMetricSpec:
              metricSpecName: "{{ metricSpecName }}"
              metricSpecParameters: "{{ metricSpecParameters }}"
            metadata:
              otherMetadata: "{{ otherMetadata }}"
              title: "{{ title }}"
              scoreRange:
                min: {{ min }}
                step: {{ step }}
                description: "{{ description }}"
                max: {{ max }}
            customCodeExecutionSpec:
              evaluationFunction: "{{ evaluationFunction }}"
            pairwiseMetricSpec:
              systemInstruction: "{{ systemInstruction }}"
              baselineResponseFieldName: "{{ baselineResponseFieldName }}"
              candidateResponseFieldName: "{{ candidateResponseFieldName }}"
              metricPromptTemplate: "{{ metricPromptTemplate }}"
              customOutputFormatConfig:
                returnRawOutput: {{ returnRawOutput }}
            aggregationMetrics: "{{ aggregationMetrics }}"
            llmBasedMetricSpec:
              systemInstruction: "{{ systemInstruction }}"
              judgeAutoraterConfig:
                samplingCount: {{ samplingCount }}
                flipEnabled: {{ flipEnabled }}
                autoraterModel: "{{ autoraterModel }}"
                generationConfig: "{{ generationConfig }}"
              rubricGroupKey: "{{ rubricGroupKey }}"
              resultParserConfig:
                customCodeParserConfig: "{{ customCodeParserConfig }}"
              predefinedRubricGenerationSpec:
                metricSpecName: "{{ metricSpecName }}"
                metricSpecParameters: "{{ metricSpecParameters }}"
              metricPromptTemplate: "{{ metricPromptTemplate }}"
              rubricGenerationSpec:
                promptTemplate: "{{ promptTemplate }}"
                modelConfig: "{{ modelConfig }}"
                rubricTypeOntology: "{{ rubricTypeOntology }}"
                rubricContentType: "{{ rubricContentType }}"
              additionalConfig: "{{ additionalConfig }}"
        inferenceGenerationConfig:
          speechConfig:
            languageCode: "{{ languageCode }}"
            multiSpeakerVoiceConfig:
              speakerVoiceConfigs: "{{ speakerVoiceConfigs }}"
            voiceConfig:
              replicatedVoiceConfig: "{{ replicatedVoiceConfig }}"
              prebuiltVoiceConfig: "{{ prebuiltVoiceConfig }}"
          seed: {{ seed }}
          imageConfig:
            personGeneration: "{{ personGeneration }}"
            aspectRatio: "{{ aspectRatio }}"
            imageSize: "{{ imageSize }}"
            prominentPeople: "{{ prominentPeople }}"
            imageOutputOptions:
              compressionQuality: {{ compressionQuality }}
              mimeType: "{{ mimeType }}"
          audioTranscriptionConfig:
            customVocabulary:
              - "{{ customVocabulary }}"
            languageHints:
              languageCodes: "{{ languageCodes }}"
            diarization: {{ diarization }}
            adaptationPhrases:
              - "{{ adaptationPhrases }}"
            languageAuto: "{{ languageAuto }}"
            wordTimestamp: {{ wordTimestamp }}
          maxOutputTokens: {{ maxOutputTokens }}
          responseModalities:
            - "{{ responseModalities }}"
          topP: {{ topP }}
          responseJsonSchema: "{{ responseJsonSchema }}"
          responseFormat:
            - audio:
                mimeType: "{{ mimeType }}"
                bitRate: {{ bitRate }}
                delivery: "{{ delivery }}"
                sampleRate: {{ sampleRate }}
              text:
                mimeType: "{{ mimeType }}"
                schema: "{{ schema }}"
              image:
                mimeType: "{{ mimeType }}"
                delivery: "{{ delivery }}"
                aspectRatio: "{{ aspectRatio }}"
                imageSize: "{{ imageSize }}"
              video:
                delivery: "{{ delivery }}"
                aspectRatio: "{{ aspectRatio }}"
                gcsUri: "{{ gcsUri }}"
                duration: "{{ duration }}"
          thinkingConfig:
            includeThoughts: {{ includeThoughts }}
            thinkingLevel: "{{ thinkingLevel }}"
            thinkingBudget: {{ thinkingBudget }}
          enableAffectiveDialog: {{ enableAffectiveDialog }}
          topK: {{ topK }}
          logprobs: {{ logprobs }}
          presencePenalty: {{ presencePenalty }}
          candidateCount: {{ candidateCount }}
          responseSchema:
            ref: "{{ ref }}"
            enum:
              - "{{ enum }}"
            minLength: "{{ minLength }}"
            title: "{{ title }}"
            minimum: {{ minimum }}
            maxProperties: "{{ maxProperties }}"
            properties: "{{ properties }}"
            nullable: {{ nullable }}
            example: "{{ example }}"
            minProperties: "{{ minProperties }}"
            type: "{{ type }}"
            pattern: "{{ pattern }}"
            additionalProperties: "{{ additionalProperties }}"
            format: "{{ format }}"
            propertyOrdering:
              - "{{ propertyOrdering }}"
            minItems: "{{ minItems }}"
            maximum: {{ maximum }}
            maxLength: "{{ maxLength }}"
            items:
              ref: "{{ ref }}"
              enum: "{{ enum }}"
              minLength: "{{ minLength }}"
              title: "{{ title }}"
              minimum: {{ minimum }}
              maxProperties: "{{ maxProperties }}"
              properties: "{{ properties }}"
              nullable: {{ nullable }}
              example: "{{ example }}"
              minProperties: "{{ minProperties }}"
              type: "{{ type }}"
              pattern: "{{ pattern }}"
              additionalProperties: "{{ additionalProperties }}"
              format: "{{ format }}"
              propertyOrdering: "{{ propertyOrdering }}"
              minItems: "{{ minItems }}"
              maximum: {{ maximum }}
              maxLength: "{{ maxLength }}"
              items: "{{ items }}"
              description: "{{ description }}"
              required: "{{ required }}"
              maxItems: "{{ maxItems }}"
              defs: "{{ defs }}"
              anyOf: "{{ anyOf }}"
              default: "{{ default }}"
            description: "{{ description }}"
            required:
              - "{{ required }}"
            maxItems: "{{ maxItems }}"
            defs: "{{ defs }}"
            anyOf:
              - ref: "{{ ref }}"
                enum: "{{ enum }}"
                minLength: "{{ minLength }}"
                title: "{{ title }}"
                minimum: {{ minimum }}
                maxProperties: "{{ maxProperties }}"
                properties: "{{ properties }}"
                nullable: {{ nullable }}
                example: "{{ example }}"
                minProperties: "{{ minProperties }}"
                type: "{{ type }}"
                pattern: "{{ pattern }}"
                additionalProperties: "{{ additionalProperties }}"
                format: "{{ format }}"
                propertyOrdering: "{{ propertyOrdering }}"
                minItems: "{{ minItems }}"
                maximum: {{ maximum }}
                maxLength: "{{ maxLength }}"
                items:
                  ref: "{{ ref }}"
                  enum: "{{ enum }}"
                  minLength: "{{ minLength }}"
                  title: "{{ title }}"
                  minimum: {{ minimum }}
                  maxProperties: "{{ maxProperties }}"
                  properties: "{{ properties }}"
                  nullable: {{ nullable }}
                  example: "{{ example }}"
                  minProperties: "{{ minProperties }}"
                  type: "{{ type }}"
                  pattern: "{{ pattern }}"
                  additionalProperties: "{{ additionalProperties }}"
                  format: "{{ format }}"
                  propertyOrdering: "{{ propertyOrdering }}"
                  minItems: "{{ minItems }}"
                  maximum: {{ maximum }}
                  maxLength: "{{ maxLength }}"
                  items: "{{ items }}"
                  description: "{{ description }}"
                  required: "{{ required }}"
                  maxItems: "{{ maxItems }}"
                  defs: "{{ defs }}"
                  anyOf: "{{ anyOf }}"
                  default: "{{ default }}"
                description: "{{ description }}"
                required: "{{ required }}"
                maxItems: "{{ maxItems }}"
                defs: "{{ defs }}"
                anyOf: "{{ anyOf }}"
                default: "{{ default }}"
            default: "{{ default }}"
          routingConfig:
            manualMode:
              modelName: "{{ modelName }}"
            autoMode:
              modelRoutingPreference: "{{ modelRoutingPreference }}"
          audioTimestamp: {{ audioTimestamp }}
          temperature: {{ temperature }}
          stopSequences:
            - "{{ stopSequences }}"
          responseLogprobs: {{ responseLogprobs }}
          responseMimeType: "{{ responseMimeType }}"
          mediaResolution: "{{ mediaResolution }}"
          frequencyPenalty: {{ frequencyPenalty }}
        autoraterConfig:
          samplingCount: {{ samplingCount }}
          flipEnabled: {{ flipEnabled }}
          autoraterModel: "{{ autoraterModel }}"
          generationConfig:
            speechConfig:
              languageCode: "{{ languageCode }}"
              multiSpeakerVoiceConfig: "{{ multiSpeakerVoiceConfig }}"
              voiceConfig: "{{ voiceConfig }}"
            seed: {{ seed }}
            imageConfig:
              personGeneration: "{{ personGeneration }}"
              aspectRatio: "{{ aspectRatio }}"
              imageSize: "{{ imageSize }}"
              prominentPeople: "{{ prominentPeople }}"
              imageOutputOptions: "{{ imageOutputOptions }}"
            audioTranscriptionConfig:
              customVocabulary: "{{ customVocabulary }}"
              languageHints: "{{ languageHints }}"
              diarization: {{ diarization }}
              adaptationPhrases: "{{ adaptationPhrases }}"
              languageAuto: "{{ languageAuto }}"
              wordTimestamp: {{ wordTimestamp }}
            maxOutputTokens: {{ maxOutputTokens }}
            responseModalities:
              - "{{ responseModalities }}"
            topP: {{ topP }}
            responseJsonSchema: "{{ responseJsonSchema }}"
            responseFormat:
              - audio:
                  mimeType: "{{ mimeType }}"
                  bitRate: {{ bitRate }}
                  delivery: "{{ delivery }}"
                  sampleRate: {{ sampleRate }}
                text:
                  mimeType: "{{ mimeType }}"
                  schema: "{{ schema }}"
                image:
                  mimeType: "{{ mimeType }}"
                  delivery: "{{ delivery }}"
                  aspectRatio: "{{ aspectRatio }}"
                  imageSize: "{{ imageSize }}"
                video:
                  delivery: "{{ delivery }}"
                  aspectRatio: "{{ aspectRatio }}"
                  gcsUri: "{{ gcsUri }}"
                  duration: "{{ duration }}"
            thinkingConfig:
              includeThoughts: {{ includeThoughts }}
              thinkingLevel: "{{ thinkingLevel }}"
              thinkingBudget: {{ thinkingBudget }}
            enableAffectiveDialog: {{ enableAffectiveDialog }}
            topK: {{ topK }}
            logprobs: {{ logprobs }}
            presencePenalty: {{ presencePenalty }}
            candidateCount: {{ candidateCount }}
            responseSchema:
              ref: "{{ ref }}"
              enum: "{{ enum }}"
              minLength: "{{ minLength }}"
              title: "{{ title }}"
              minimum: {{ minimum }}
              maxProperties: "{{ maxProperties }}"
              properties: "{{ properties }}"
              nullable: {{ nullable }}
              example: "{{ example }}"
              minProperties: "{{ minProperties }}"
              type: "{{ type }}"
              pattern: "{{ pattern }}"
              additionalProperties: "{{ additionalProperties }}"
              format: "{{ format }}"
              propertyOrdering: "{{ propertyOrdering }}"
              minItems: "{{ minItems }}"
              maximum: {{ maximum }}
              maxLength: "{{ maxLength }}"
              items: "{{ items }}"
              description: "{{ description }}"
              required: "{{ required }}"
              maxItems: "{{ maxItems }}"
              defs: "{{ defs }}"
              anyOf: "{{ anyOf }}"
              default: "{{ default }}"
            routingConfig:
              manualMode: "{{ manualMode }}"
              autoMode: "{{ autoMode }}"
            audioTimestamp: {{ audioTimestamp }}
            temperature: {{ temperature }}
            stopSequences:
              - "{{ stopSequences }}"
            responseLogprobs: {{ responseLogprobs }}
            responseMimeType: "{{ responseMimeType }}"
            mediaResolution: "{{ mediaResolution }}"
            frequencyPenalty: {{ frequencyPenalty }}
        outputConfig:
          gcsDestination:
            outputUriPrefix: "{{ outputUriPrefix }}"
        datasetCustomMetrics:
          - displayName: "{{ displayName }}"
            aggregationFunction: "{{ aggregationFunction }}"
      hyperParameters:
        adapterSize: "{{ adapterSize }}"
        epochCount: "{{ epochCount }}"
        learningRateMultiplier: {{ learningRateMultiplier }}
  - name: encryptionSpec
    description: |
      Customer-managed encryption key options for a TuningJob. If this is set, then all resources created by the TuningJob will be encrypted with the provided encryption key.
    value:
      kmsKeyName: "{{ kmsKeyName }}"
  - name: preferenceOptimizationSpec
    description: |
      Tuning Spec for Preference Optimization.
    value:
      hyperParameters:
        adapterSize: "{{ adapterSize }}"
        beta: {{ beta }}
        epochCount: "{{ epochCount }}"
        learningRateMultiplier: {{ learningRateMultiplier }}
      validationDatasetUri: "{{ validationDatasetUri }}"
      trainingDatasetUri: "{{ trainingDatasetUri }}"
      exportLastCheckpointOnly: {{ exportLastCheckpointOnly }}
  - name: preTunedModel
    description: |
      The pre-tuned model for continuous tuning.
    value:
      tunedModelName: "{{ tunedModelName }}"
      checkpointId: "{{ checkpointId }}"
      baseModel: "{{ baseModel }}"
  - name: serviceAccount
    value: "{{ serviceAccount }}"
    description: |
      The service account that the tuningJob workload runs as. If not specified, the Vertex AI Secure Fine-Tuned Service Agent in the project will be used. See https://cloud.google.com/iam/docs/service-agents#vertex-ai-secure-fine-tuning-service-agent Users starting the pipeline must have the `iam.serviceAccounts.actAs` permission on this service account.

Lifecycle Methods

rebase_tuned_model

Rebase a tuned model. A rebase operation takes a model that was previously tuned on a base model version, and retunes it on a new base model version. The rebase operation creates a new tuning job and a new tuned model.

EXEC google.aiplatform.tuning_jobs.rebase_tuned_model 
@projectsId='{{ projectsId }}' --required, 
@locationsId='{{ locationsId }}' --required 
@@json=
'{
"tuningJob": "{{ tuningJob }}", 
"deployToSameEndpoint": {{ deployToSameEndpoint }}, 
"tunedModelRef": "{{ tunedModelRef }}", 
"artifactDestination": "{{ artifactDestination }}"
}'
;

cancel

Cancels a tuning job. Starts an asynchronous cancellation request. The server makes a best effort to cancel the job, but success is not guaranteed. Clients can use GenAiTuningService.GetTuningJob or other methods to check whether the cancellation succeeded or whether the job completed despite cancellation. On successful cancellation, the tuning job is not deleted. Instead, its state is set to CANCELLED, and error is set to a status with a google.rpc.Status.code of 1, corresponding to Code.CANCELLED.

EXEC google.aiplatform.tuning_jobs.cancel 
@projectsId='{{ projectsId }}' --required, 
@locationsId='{{ locationsId }}' --required, 
@tuningJobsId='{{ tuningJobsId }}' --required
;