Project googleapis/googleapis

optional BleuSpec metric_spec = 1
Required. Spec for bleu score metric.
repeated BleuInstance instances = 2
Required. Repeated bleu instances.

Spec for bleu instance.

Used in: BleuInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Required. Ground truth used to compare against the prediction.

Bleu metric value for an instance.

Used in: BleuResults

optional float score = 1
Output only. Bleu score.

Results for bleu metric.

repeated BleuMetricValue bleu_metric_values = 1
Output only. Bleu metric values.

Spec for bleu score metric - calculates the precision of n-grams in the prediction as compared to reference - returns a score ranging between 0 to 1.

Used in: BleuInput

bool use_effective_order = 1
Optional. Whether to use_effective_order to compute bleu score.

Content blob. It's preferred to send as [text][google.cloud.aiplatform.v1.Part.text] directly rather than raw bytes.

Used in: Part

string mime_type = 1
Required. The IANA standard MIME type of the source data.
bytes data = 2
Required. Raw bytes.

Config for blur baseline. When enabled, a linear path from the maximally blurred image to the input image is created. Using a blurred baseline instead of zero (black image) is motivated by the BlurIG approach explained here: https://arxiv.org/abs/2004.03383

Used in: IntegratedGradientsAttribution, XraiAttribution

float max_blur_sigma = 1
The standard deviation of the blur kernel for the blurred baseline. The same blurring parameter is used for both the height and the width dimension. If not set, the method defaults to the zero (i.e. black for images) baseline.

A list of boolean values.

Used in: FeatureValue

repeated bool values = 1
A list of bool values.

A resource used in LLM queries for users to explicitly specify what to cache and how to cache.

Used as response type in: GenAiCacheService.CreateCachedContent, GenAiCacheService.GetCachedContent, GenAiCacheService.UpdateCachedContent

Used as field type in: CreateCachedContentRequest, ListCachedContentsResponse, UpdateCachedContentRequest

oneof expiration
Expiration time of the cached content.
- protobuf.Timestamp expire_time = 9
  Timestamp of when this resource is considered expired. This is *always* provided on output, regardless of what was sent on input.
- protobuf.Duration ttl = 10
  Input only. The TTL for this resource. The expiration time is computed: now + TTL.
string name = 1
Immutable. Identifier. The server-generated resource name of the cached content Format: projects/{project}/locations/{location}/cachedContents/{cached_content}
string display_name = 11
Optional. Immutable. The user-generated meaningful display name of the cached content.
string model = 2
Immutable. The name of the `Model` to use for cached content. Currently, only the published Gemini base models are supported, in form of projects/{PROJECT}/locations/{LOCATION}/publishers/google/models/{MODEL}
optional Content system_instruction = 3
Optional. Input only. Immutable. Developer set system instruction. Currently, text only
repeated Content contents = 4
Optional. Input only. Immutable. The content to cache
repeated Tool tools = 5
Optional. Input only. Immutable. A list of `Tools` the model may use to generate the next response
optional ToolConfig tool_config = 6
Optional. Input only. Immutable. Tool config. This config is shared for all tools
optional protobuf.Timestamp create_time = 7
Output only. Creation time of the cache entry.
optional protobuf.Timestamp update_time = 8
Output only. When the cache entry was last updated in UTC time.
optional CachedContent.UsageMetadata usage_metadata = 12
Output only. Metadata on the usage of the cached content.
optional EncryptionSpec encryption_spec = 13
Input only. Immutable. Customer-managed encryption key spec for a `CachedContent`. If set, this `CachedContent` and all its sub-resources will be secured by this key.

Metadata on the usage of the cached content.

Used in: CachedContent

int32 total_token_count = 1
Total number of tokens that the cached content consumes.
int32 text_count = 2
Number of text characters.
int32 image_count = 3
Number of images.
int32 video_duration_seconds = 4
Duration of video in seconds.
int32 audio_duration_seconds = 5
Duration of audio in seconds.

A response candidate generated from the model.

Used in: GenerateContentResponse

int32 index = 1
Output only. Index of the candidate.
optional Content content = 2
Output only. Content parts of the candidate.
double score = 8
Output only. Confidence score of the candidate.
double avg_logprobs = 9
Output only. Average log probability score of the candidate.
optional LogprobsResult logprobs_result = 10
Output only. Log-likelihood scores for the response tokens and top tokens
Candidate.FinishReason finish_reason = 3
Output only. The reason why the model stopped generating tokens. If empty, the model has not stopped generating the tokens.
repeated SafetyRating safety_ratings = 4
Output only. List of ratings for the safety of a response candidate. There is at most one rating per category.
optional string finish_message = 5
Output only. Describes the reason the mode stopped generating tokens in more detail. This is only filled when `finish_reason` is set.
optional CitationMetadata citation_metadata = 6
Output only. Source attribution of the generated content.
optional GroundingMetadata grounding_metadata = 7
Output only. Metadata specifies sources used to ground generated content.

The reason why the model stopped generating tokens. If empty, the model has not stopped generating the tokens.

Used in: Candidate

FINISH_REASON_UNSPECIFIED = 0
The finish reason is unspecified.
STOP = 1
Token generation reached a natural stopping point or a configured stop sequence.
MAX_TOKENS = 2
Token generation reached the configured maximum output tokens.
SAFETY = 3
Token generation stopped because the content potentially contains safety violations. NOTE: When streaming, [content][google.cloud.aiplatform.v1.Candidate.content] is empty if content filters blocks the output.
RECITATION = 4
Token generation stopped because the content potentially contains copyright violations.
OTHER = 5
All other reasons that stopped the token generation.
BLOCKLIST = 6
Token generation stopped because the content contains forbidden terms.
PROHIBITED_CONTENT = 7
Token generation stopped for potentially containing prohibited content.
SPII = 8
Token generation stopped because the content potentially contains Sensitive Personally Identifiable Information (SPII).
MALFORMED_FUNCTION_CALL = 9
The function call generated by the model is invalid.

This message will be placed in the metadata field of a google.longrunning.Operation associated with a CheckTrialEarlyStoppingState request.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for suggesting Trials.
string study = 2
The name of the Study that the Trial belongs to.
string trial = 3
The Trial name.

Response message for [VizierService.CheckTrialEarlyStoppingState][google.cloud.aiplatform.v1.VizierService.CheckTrialEarlyStoppingState].

bool should_stop = 1
True if the Trial should stop.

Describes the machine learning model version checkpoint.

Used in: Model

string checkpoint_id = 1
The ID of the checkpoint.
int64 epoch = 2
The epoch of the checkpoint.
int64 step = 3
The step of the checkpoint.

Source attributions for content.

Used in: CitationMetadata

int32 start_index = 1
Output only. Start index into the content.
int32 end_index = 2
Output only. End index into the content.
string uri = 3
Output only. Url reference of the attribution.
string title = 4
Output only. Title of the attribution.
string license = 5
Output only. License of the attribution.
optional type.Date publication_date = 6
Output only. Publication date of the attribution.

A collection of source attributions for a piece of content.

Used in: Candidate

repeated Citation citations = 1
Output only. List of citations.

Claim that is extracted from the input text and facts that support it.

Used in: CorroborateContentResponse

optional int32 start_index = 1
Index in the input text where the claim starts (inclusive).
optional int32 end_index = 2
Index in the input text where the claim ends (exclusive).
repeated int32 fact_indexes = 3
Indexes of the facts supporting this claim.
optional float score = 4
Confidence score of this corroboration.

Configurations (e.g. inference timeout) that are applied on your endpoints.

Used in: Endpoint

optional protobuf.Duration inference_timeout = 1
Customizable online prediction request timeout.

Result of executing the [ExecutableCode]. Always follows a `part` containing the [ExecutableCode].

Used in: Part

CodeExecutionResult.Outcome outcome = 1
Required. Outcome of the code execution.
string output = 2
Optional. Contains stdout when code execution is successful, stderr or other description otherwise.

Enumeration of possible outcomes of the code execution.

Used in: CodeExecutionResult

OUTCOME_UNSPECIFIED = 0
Unspecified status. This value should not be used.
OUTCOME_OK = 1
Code execution completed successfully.
OUTCOME_FAILED = 2
Code execution finished but with a failure. `stderr` should contain the reason.
OUTCOME_DEADLINE_EXCEEDED = 3
Code execution ran for too long, and was cancelled. There may or may not be a partial output present.

Input for coherence metric.

optional CoherenceSpec metric_spec = 1
Required. Spec for coherence score metric.
optional CoherenceInstance instance = 2
Required. Coherence instance.

Spec for coherence instance.

Used in: CoherenceInput

optional string prediction = 1
Required. Output of the evaluated model.

Spec for coherence result.

optional float score = 1
Output only. Coherence score.
string explanation = 2
Output only. Explanation for coherence score.
optional float confidence = 3
Output only. Confidence for coherence score.

Spec for coherence score metric.

Used in: CoherenceInput

int32 version = 1
Optional. Which version to use for evaluation.

Input for Comet metric.

optional CometSpec metric_spec = 1
Required. Spec for comet metric.
optional CometInstance instance = 2
Required. Comet instance.

Spec for Comet instance - The fields used for evaluation are dependent on the comet version.

Used in: CometInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Optional. Ground truth used to compare against the prediction.
optional string source = 3
Optional. Source text in original language.

Spec for Comet result - calculates the comet score for the given instance using the version specified in the spec.

optional float score = 1
Output only. Comet score. Range depends on version.

Spec for Comet metric.

Used in: CometInput

optional CometSpec.CometVersion version = 1
Required. Which version to use for evaluation.
string source_language = 2
Optional. Source language in BCP-47 format.
string target_language = 3
Optional. Target language in BCP-47 format. Covers both prediction and reference.

Comet version options.

Used in: CometSpec

COMET_VERSION_UNSPECIFIED = 0
Comet version unspecified.
COMET_22_SRC_REF = 2
Comet 22 for translation + source + reference (source-reference-combined).

Success and error statistics of processing multiple entities (for example, DataItems or structured data rows) in batch.

Used in: BatchPredictionJob

int64 successful_count = 1
Output only. The number of entities that had been processed successfully.
int64 failed_count = 2
Output only. The number of entities for which any error was encountered.
int64 incomplete_count = 3
Output only. In cases when enough errors are encountered a job, pipeline, or operation may be failed as a whole. Below is the number of entities for which the processing had not been finished (either in successful or failed state). Set to -1 if the number is unknown (for example, the operation failed before the total entity number could be collected).
int64 successful_forecast_point_count = 5
Output only. The number of the successful forecast points that are generated by the forecasting model. This is ONLY used by the forecasting batch prediction.

The Container Registry location for the container image.

Used in: ExportModelRequest.OutputConfig

string output_uri = 1
Required. Container Registry URI of a container image. Only Google Container Registry and Artifact Registry are supported now. Accepted forms: * Google Container Registry path. For example: `gcr.io/projectId/imageName:tag`. * Artifact Registry path. For example: `us-central1-docker.pkg.dev/projectId/repoName/imageName:tag`. If a tag is not specified, "latest" will be used as the default tag.

The spec of a Container.

Used in: WorkerPoolSpec

string image_uri = 1
Required. The URI of a container image in the Container Registry that is to be run on each worker replica.
repeated string command = 2
The command to be invoked when the container is started. It overrides the entrypoint instruction in Dockerfile when provided.
repeated string args = 3
The arguments to be passed when starting the container.
repeated EnvVar env = 4
Environment variables to be passed to the container. Maximum limit is 100.

The base structured datatype containing multi-part content of a message. A `Content` includes a `role` field designating the producer of the `Content` and a `parts` field containing multi-part data that contains the content of the message turn.

Used in: AugmentPromptRequest, AugmentPromptResponse, CachedContent, Candidate, ComputeTokensRequest, CorroborateContentRequest, CountTokensRequest, GenerateContentRequest, SupervisedTuningDataStats

string role = 1
Optional. The producer of the content. Must be either 'user' or 'model'. Useful to set for multi-turn conversations, otherwise can be left blank or unset.
repeated Part parts = 2
Required. Ordered `Parts` that constitute a single message. Parts may have different IANA MIME types.

Instance of a general context.

Used as response type in: MetadataService.CreateContext, MetadataService.GetContext, MetadataService.UpdateContext

Used as field type in: CreateContextRequest, ListContextsResponse, PipelineJobDetail, UpdateContextRequest

string name = 1
Immutable. The resource name of the Context.
string display_name = 2
User provided display name of the Context. May be up to 128 Unicode characters.
string etag = 8
An eTag used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 9
The labels with user-defined metadata to organize your Contexts. 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. No more than 64 user labels can be associated with one Context (System labels are excluded).
optional protobuf.Timestamp create_time = 10
Output only. Timestamp when this Context was created.
optional protobuf.Timestamp update_time = 11
Output only. Timestamp when this Context was last updated.
repeated string parent_contexts = 12
Output only. A list of resource names of Contexts that are parents of this Context. A Context may have at most 10 parent_contexts.
string schema_title = 13
The title of the schema describing the metadata. Schema title and version is expected to be registered in earlier Create Schema calls. And both are used together as unique identifiers to identify schemas within the local metadata store.
string schema_version = 14
The version of the schema in schema_name to use. Schema title and version is expected to be registered in earlier Create Schema calls. And both are used together as unique identifiers to identify schemas within the local metadata store.
optional protobuf.Struct metadata = 15
Properties of the Context. Top level metadata keys' heading and trailing spaces will be trimmed. The size of this field should not exceed 200KB.
string description = 16
Description of the Context

Details of [ModelService.CopyModel][google.cloud.aiplatform.v1.ModelService.CopyModel] operation.

optional GenericOperationMetadata generic_metadata = 1
The common part of the operation metadata.

Response message of [ModelService.CopyModel][google.cloud.aiplatform.v1.ModelService.CopyModel] operation.

string model = 1
The name of the copied Model resource. Format: `projects/{project}/locations/{location}/models/{model}`
string model_version_id = 2
Output only. The version ID of the model that is copied.

RagCorpus status.

Used in: RagCorpus

CorpusStatus.State state = 1
Output only. RagCorpus life state.
string error_status = 2
Output only. Only when the `state` field is ERROR.

RagCorpus life state.

Used in: CorpusStatus

UNKNOWN = 0
This state is not supposed to happen.
INITIALIZED = 1
RagCorpus resource entry is initialized, but hasn't done validation.
ACTIVE = 2
RagCorpus is provisioned successfully and is ready to serve.
ERROR = 3
RagCorpus is in a problematic situation. See `error_message` field for details.

Parameters that can be overrided per request.

Used in: CorroborateContentRequest

double citation_threshold = 1
Optional. Only return claims with citation score larger than the threshold.

Runtime operation information for [DatasetService.CreateDataset][google.cloud.aiplatform.v1.DatasetService.CreateDataset].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Runtime operation information for [DatasetService.CreateDatasetVersion][google.cloud.aiplatform.v1.DatasetService.CreateDatasetVersion].

optional GenericOperationMetadata generic_metadata = 1
The common part of the operation metadata.

Runtime operation information for CreateDeploymentResourcePool method.

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Runtime operation information for [EndpointService.CreateEndpoint][google.cloud.aiplatform.v1.EndpointService.CreateEndpoint].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Details of operations that perform create EntityType.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for EntityType.

Details of operations that perform create FeatureGroup.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for FeatureGroup.

Details of operations that perform create FeatureOnlineStore.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for FeatureOnlineStore.

Details of operations that perform create Feature.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for Feature.

Request message for [FeaturestoreService.CreateFeature][google.cloud.aiplatform.v1.FeaturestoreService.CreateFeature]. Request message for [FeatureRegistryService.CreateFeature][google.cloud.aiplatform.v1.FeatureRegistryService.CreateFeature].

Used as request type in: FeatureRegistryService.CreateFeature, FeaturestoreService.CreateFeature

Used as field type in: BatchCreateFeaturesRequest

string parent = 1
Required. The resource name of the EntityType or FeatureGroup to create a Feature. Format for entity_type as parent: `projects/{project}/locations/{location}/featurestores/{featurestore}/entityTypes/{entity_type}` Format for feature_group as parent: `projects/{project}/locations/{location}/featureGroups/{feature_group}`
optional Feature feature = 2
Required. The Feature to create.
string feature_id = 3
Required. The ID to use for the Feature, which will become the final component of the Feature's resource name. This value may be up to 128 characters, and valid characters are `[a-z0-9_]`. The first character cannot be a number. The value must be unique within an EntityType/FeatureGroup.

Details of operations that perform create FeatureView.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for FeatureView Create.

Details of operations that perform create Featurestore.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for Featurestore.

Runtime operation information for [IndexEndpointService.CreateIndexEndpoint][google.cloud.aiplatform.v1.IndexEndpointService.CreateIndexEndpoint].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Runtime operation information for [IndexService.CreateIndex][google.cloud.aiplatform.v1.IndexService.CreateIndex].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.
optional NearestNeighborSearchOperationMetadata nearest_neighbor_search_operation_metadata = 2
The operation metadata with regard to Matching Engine Index operation.

Details of operations that perform [MetadataService.CreateMetadataStore][google.cloud.aiplatform.v1.MetadataService.CreateMetadataStore].

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for creating a MetadataStore.

Metadata information for [NotebookService.CreateNotebookExecutionJob][google.cloud.aiplatform.v1.NotebookService.CreateNotebookExecutionJob].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.
string progress_message = 2
A human-readable message that shows the intermediate progress details of NotebookRuntime.

Request message for [NotebookService.CreateNotebookExecutionJob]

Used as request type in: NotebookService.CreateNotebookExecutionJob

Used as field type in: Schedule

string parent = 1
Required. The resource name of the Location to create the NotebookExecutionJob. Format: `projects/{project}/locations/{location}`
optional NotebookExecutionJob notebook_execution_job = 2
Required. The NotebookExecutionJob to create.
string notebook_execution_job_id = 3
Optional. User specified ID for the NotebookExecutionJob.

Metadata information for [NotebookService.CreateNotebookRuntimeTemplate][google.cloud.aiplatform.v1.NotebookService.CreateNotebookRuntimeTemplate].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Details of operations that perform create PersistentResource.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for PersistentResource.
string progress_message = 2
Progress Message for Create LRO

Request message for [PipelineService.CreatePipelineJob][google.cloud.aiplatform.v1.PipelineService.CreatePipelineJob].

Used as request type in: PipelineService.CreatePipelineJob

Used as field type in: Schedule

string parent = 1
Required. The resource name of the Location to create the PipelineJob in. Format: `projects/{project}/locations/{location}`
optional PipelineJob pipeline_job = 2
Required. The PipelineJob to create.
string pipeline_job_id = 3
The ID to use for the PipelineJob, which will become the final component of the PipelineJob name. If not provided, an ID will be automatically generated. This value should be less than 128 characters, and valid characters are `/[a-z][0-9]-/`.

Runtime operation information for [VertexRagDataService.CreateRagCorpus][google.cloud.aiplatform.v1.VertexRagDataService.CreateRagCorpus].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Details of [ReasoningEngineService.CreateReasoningEngine][google.cloud.aiplatform.v1.ReasoningEngineService.CreateReasoningEngine] operation.

optional GenericOperationMetadata generic_metadata = 1
The common part of the operation metadata.

Details of operations that perform create FeatureGroup.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for Feature.

Runtime operation information for [SpecialistPoolService.CreateSpecialistPool][google.cloud.aiplatform.v1.SpecialistPoolService.CreateSpecialistPool].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Details of operations that perform create Tensorboard.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for Tensorboard.

Request message for [TensorboardService.CreateTensorboardRun][google.cloud.aiplatform.v1.TensorboardService.CreateTensorboardRun].

Used as request type in: TensorboardService.CreateTensorboardRun

Used as field type in: BatchCreateTensorboardRunsRequest

string parent = 1
Required. The resource name of the TensorboardExperiment to create the TensorboardRun in. Format: `projects/{project}/locations/{location}/tensorboards/{tensorboard}/experiments/{experiment}`
optional TensorboardRun tensorboard_run = 2
Required. The TensorboardRun to create.
string tensorboard_run_id = 3
Required. The ID to use for the Tensorboard run, which becomes the final component of the Tensorboard run's resource name. This value should be 1-128 characters, and valid characters are `/[a-z][0-9]-/`.

Request message for [TensorboardService.CreateTensorboardTimeSeries][google.cloud.aiplatform.v1.TensorboardService.CreateTensorboardTimeSeries].

Used as request type in: TensorboardService.CreateTensorboardTimeSeries

Used as field type in: BatchCreateTensorboardTimeSeriesRequest

string parent = 1
Required. The resource name of the TensorboardRun to create the TensorboardTimeSeries in. Format: `projects/{project}/locations/{location}/tensorboards/{tensorboard}/experiments/{experiment}/runs/{run}`
string tensorboard_time_series_id = 3
Optional. The user specified unique ID to use for the TensorboardTimeSeries, which becomes the final component of the TensorboardTimeSeries's resource name. This value should match "[a-z0-9][a-z0-9-]{0, 127}"
optional TensorboardTimeSeries tensorboard_time_series = 2
Required. The TensorboardTimeSeries to create.

The storage details for CSV output content.

Used in: FeatureValueDestination

optional GcsDestination gcs_destination = 1
Required. Google Cloud Storage location.

The storage details for CSV input content.

Used in: BatchReadFeatureValuesRequest, EntityIdSelector, ImportFeatureValuesRequest

optional GcsSource gcs_source = 1
Required. Google Cloud Storage location.

Represents a job that runs custom workloads such as a Docker container or a Python package. A CustomJob can have multiple worker pools and each worker pool can have its own machine and input spec. A CustomJob will be cleaned up once the job enters terminal state (failed or succeeded).

Used as response type in: JobService.CreateCustomJob, JobService.GetCustomJob

Used as field type in: CreateCustomJobRequest, ListCustomJobsResponse

string name = 1
Output only. Resource name of a CustomJob.
string display_name = 2
Required. The display name of the CustomJob. The name can be up to 128 characters long and can consist of any UTF-8 characters.
optional CustomJobSpec job_spec = 4
Required. Job spec.
JobState state = 5
Output only. The detailed state of the job.
optional protobuf.Timestamp create_time = 6
Output only. Time when the CustomJob was created.
optional protobuf.Timestamp start_time = 7
Output only. Time when the CustomJob for the first time entered the `JOB_STATE_RUNNING` state.
optional protobuf.Timestamp end_time = 8
Output only. Time when the CustomJob entered any of the following states: `JOB_STATE_SUCCEEDED`, `JOB_STATE_FAILED`, `JOB_STATE_CANCELLED`.
optional protobuf.Timestamp update_time = 9
Output only. Time when the CustomJob was most recently updated.
optional rpc.Status error = 10
Output only. Only populated when job's state is `JOB_STATE_FAILED` or `JOB_STATE_CANCELLED`.
map<string, string> labels = 11
The labels with user-defined metadata to organize CustomJobs. 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.
optional EncryptionSpec encryption_spec = 12
Customer-managed encryption key options for a CustomJob. If this is set, then all resources created by the CustomJob will be encrypted with the provided encryption key.
map<string, string> web_access_uris = 16
Output only. URIs for accessing [interactive shells](https://cloud.google.com/vertex-ai/docs/training/monitor-debug-interactive-shell) (one URI for each training node). Only available if [job_spec.enable_web_access][google.cloud.aiplatform.v1.CustomJobSpec.enable_web_access] is `true`. The keys are names of each node in the training job; for example, `workerpool0-0` for the primary node, `workerpool1-0` for the first node in the second worker pool, and `workerpool1-1` for the second node in the second worker pool. The values are the URIs for each node's interactive shell.
bool satisfies_pzs = 18
Output only. Reserved for future use.
bool satisfies_pzi = 19
Output only. Reserved for future use.

Represents the spec of a CustomJob.

Used in: CustomJob, HyperparameterTuningJob, NasJobSpec.MultiTrialAlgorithmSpec.SearchTrialSpec, NasJobSpec.MultiTrialAlgorithmSpec.TrainTrialSpec

string persistent_resource_id = 14
Optional. The ID of the PersistentResource in the same Project and Location which to run If this is specified, the job will be run on existing machines held by the PersistentResource instead of on-demand short-live machines. The network and CMEK configs on the job should be consistent with those on the PersistentResource, otherwise, the job will be rejected.
repeated WorkerPoolSpec worker_pool_specs = 1
Required. The spec of the worker pools including machine type and Docker image. All worker pools except the first one are optional and can be skipped by providing an empty value.
optional Scheduling scheduling = 3
Scheduling options for a CustomJob.
string service_account = 4
Specifies the service account for workload run-as account. Users submitting jobs must have act-as permission on this run-as account. If unspecified, the [Vertex AI Custom Code Service Agent](https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) for the CustomJob's project is used.
string network = 5
Optional. The full name of the Compute Engine [network](/compute/docs/networks-and-firewalls#networks) to which the Job should be peered. For example, `projects/12345/global/networks/myVPC`. [Format](/compute/docs/reference/rest/v1/networks/insert) is of the form `projects/{project}/global/networks/{network}`. Where {project} is a project number, as in `12345`, and {network} is a network name. To specify this field, you must have already [configured VPC Network Peering for Vertex AI](https://cloud.google.com/vertex-ai/docs/general/vpc-peering). If this field is left unspecified, the job is not peered with any network.
repeated string reserved_ip_ranges = 13
Optional. A list of names for the reserved ip ranges under the VPC network that can be used for this job. If set, we will deploy the job within the provided ip ranges. Otherwise, the job will be deployed to any ip ranges under the provided VPC network. Example: ['vertex-ai-ip-range'].
optional GcsDestination base_output_directory = 6
The Cloud Storage location to store the output of this CustomJob or HyperparameterTuningJob. For HyperparameterTuningJob, the baseOutputDirectory of each child CustomJob backing a Trial is set to a subdirectory of name [id][google.cloud.aiplatform.v1.Trial.id] under its parent HyperparameterTuningJob's baseOutputDirectory. The following Vertex AI environment variables will be passed to containers or python modules when this field is set: For CustomJob: * AIP_MODEL_DIR = `<base_output_directory>/model/` * AIP_CHECKPOINT_DIR = `<base_output_directory>/checkpoints/` * AIP_TENSORBOARD_LOG_DIR = `<base_output_directory>/logs/` For CustomJob backing a Trial of HyperparameterTuningJob: * AIP_MODEL_DIR = `<base_output_directory>/<trial_id>/model/` * AIP_CHECKPOINT_DIR = `<base_output_directory>/<trial_id>/checkpoints/` * AIP_TENSORBOARD_LOG_DIR = `<base_output_directory>/<trial_id>/logs/`
string protected_artifact_location_id = 19
The ID of the location to store protected artifacts. e.g. us-central1. Populate only when the location is different than CustomJob location. List of supported locations: https://cloud.google.com/vertex-ai/docs/general/locations
string tensorboard = 7
Optional. The name of a Vertex AI [Tensorboard][google.cloud.aiplatform.v1.Tensorboard] resource to which this CustomJob will upload Tensorboard logs. Format: `projects/{project}/locations/{location}/tensorboards/{tensorboard}`
bool enable_web_access = 10
Optional. Whether you want Vertex AI to enable [interactive shell access](https://cloud.google.com/vertex-ai/docs/training/monitor-debug-interactive-shell) to training containers. If set to `true`, you can access interactive shells at the URIs given by [CustomJob.web_access_uris][google.cloud.aiplatform.v1.CustomJob.web_access_uris] or [Trial.web_access_uris][google.cloud.aiplatform.v1.Trial.web_access_uris] (within [HyperparameterTuningJob.trials][google.cloud.aiplatform.v1.HyperparameterTuningJob.trials]).
bool enable_dashboard_access = 16
Optional. Whether you want Vertex AI to enable access to the customized dashboard in training chief container. If set to `true`, you can access the dashboard at the URIs given by [CustomJob.web_access_uris][google.cloud.aiplatform.v1.CustomJob.web_access_uris] or [Trial.web_access_uris][google.cloud.aiplatform.v1.Trial.web_access_uris] (within [HyperparameterTuningJob.trials][google.cloud.aiplatform.v1.HyperparameterTuningJob.trials]).
string experiment = 17
Optional. The Experiment associated with this job. Format: `projects/{project}/locations/{location}/metadataStores/{metadataStores}/contexts/{experiment-name}`
string experiment_run = 18
Optional. The Experiment Run associated with this job. Format: `projects/{project}/locations/{location}/metadataStores/{metadataStores}/contexts/{experiment-name}-{experiment-run-name}`
repeated string models = 20
Optional. The name of the Model resources for which to generate a mapping to artifact URIs. Applicable only to some of the Google-provided custom jobs. Format: `projects/{project}/locations/{location}/models/{model}` In order to retrieve a specific version of the model, also provide the version ID or version alias. Example: `projects/{project}/locations/{location}/models/{model}@2` or `projects/{project}/locations/{location}/models/{model}@golden` If no version ID or alias is specified, the "default" version will be returned. The "default" version alias is created for the first version of the model, and can be moved to other versions later on. There will be exactly one default version.

A piece of data in a Dataset. Could be an image, a video, a document or plain text.

Used in: DataItemView, ListDataItemsResponse

string name = 1
Output only. The resource name of the DataItem.
optional protobuf.Timestamp create_time = 2
Output only. Timestamp when this DataItem was created.
optional protobuf.Timestamp update_time = 6
Output only. Timestamp when this DataItem was last updated.
map<string, string> labels = 3
Optional. The labels with user-defined metadata to organize your DataItems. 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. No more than 64 user labels can be associated with one DataItem(System labels are excluded). See https://goo.gl/xmQnxf for more information and examples of labels. System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
optional protobuf.Value payload = 4
Required. The data that the DataItem represents (for example, an image or a text snippet). The schema of the payload is stored in the parent Dataset's [metadata schema's][google.cloud.aiplatform.v1.Dataset.metadata_schema_uri] dataItemSchemaUri field.
string etag = 7
Optional. Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
bool satisfies_pzs = 10
Output only. Reserved for future use.
bool satisfies_pzi = 11
Output only. Reserved for future use.

A container for a single DataItem and Annotations on it.

Used in: SearchDataItemsResponse

optional DataItem data_item = 1
The DataItem.
repeated Annotation annotations = 2
The Annotations on the DataItem. If too many Annotations should be returned for the DataItem, this field will be truncated per annotations_limit in request. If it was, then the has_truncated_annotations will be set to true.
bool has_truncated_annotations = 3
True if and only if the Annotations field has been truncated. It happens if more Annotations for this DataItem met the request's annotation_filter than are allowed to be returned by annotations_limit. Note that if Annotations field is not being returned due to field mask, then this field will not be set to true no matter how many Annotations are there.

DataLabelingJob is used to trigger a human labeling job on unlabeled data from the following Dataset:

Used as response type in: JobService.CreateDataLabelingJob, JobService.GetDataLabelingJob

Used as field type in: CreateDataLabelingJobRequest, ListDataLabelingJobsResponse

string name = 1
Output only. Resource name of the DataLabelingJob.
string display_name = 2
Required. The user-defined name of the DataLabelingJob. The name can be up to 128 characters long and can consist of any UTF-8 characters. Display name of a DataLabelingJob.
repeated string datasets = 3
Required. Dataset resource names. Right now we only support labeling from a single Dataset. Format: `projects/{project}/locations/{location}/datasets/{dataset}`
map<string, string> annotation_labels = 12
Labels to assign to annotations generated by this DataLabelingJob. 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. System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
int32 labeler_count = 4
Required. Number of labelers to work on each DataItem.
string instruction_uri = 5
Required. The Google Cloud Storage location of the instruction pdf. This pdf is shared with labelers, and provides detailed description on how to label DataItems in Datasets.
string inputs_schema_uri = 6
Required. Points to a YAML file stored on Google Cloud Storage describing the config for a specific type of DataLabelingJob. The schema files that can be used here are found in the https://storage.googleapis.com/google-cloud-aiplatform bucket in the /schema/datalabelingjob/inputs/ folder.
optional protobuf.Value inputs = 7
Required. Input config parameters for the DataLabelingJob.
JobState state = 8
Output only. The detailed state of the job.
int32 labeling_progress = 13
Output only. Current labeling job progress percentage scaled in interval [0, 100], indicating the percentage of DataItems that has been finished.
optional type.Money current_spend = 14
Output only. Estimated cost(in US dollars) that the DataLabelingJob has incurred to date.
optional protobuf.Timestamp create_time = 9
Output only. Timestamp when this DataLabelingJob was created.
optional protobuf.Timestamp update_time = 10
Output only. Timestamp when this DataLabelingJob was updated most recently.
optional rpc.Status error = 22
Output only. DataLabelingJob errors. It is only populated when job's state is `JOB_STATE_FAILED` or `JOB_STATE_CANCELLED`.
map<string, string> labels = 11
The labels with user-defined metadata to organize your DataLabelingJobs. 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. System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable. Following system labels exist for each DataLabelingJob: * "aiplatform.googleapis.com/schema": output only, its value is the [inputs_schema][google.cloud.aiplatform.v1.DataLabelingJob.inputs_schema_uri]'s title.
repeated string specialist_pools = 16
The SpecialistPools' resource names associated with this job.
optional EncryptionSpec encryption_spec = 20
Customer-managed encryption key spec for a DataLabelingJob. If set, this DataLabelingJob will be secured by this key. Note: Annotations created in the DataLabelingJob are associated with the EncryptionSpec of the Dataset they are exported to.
optional ActiveLearningConfig active_learning_config = 21
Parameters that configure the active learning pipeline. Active learning will label the data incrementally via several iterations. For every iteration, it will select a batch of data based on the sampling strategy.

A collection of DataItems and Annotations on them.

Used as response type in: DatasetService.GetDataset, DatasetService.UpdateDataset

Used as field type in: CreateDatasetRequest, ListDatasetsResponse, UpdateDatasetRequest

string name = 1
Output only. Identifier. The resource name of the Dataset.
string display_name = 2
Required. The user-defined name of the Dataset. The name can be up to 128 characters long and can consist of any UTF-8 characters.
string description = 16
The description of the Dataset.
string metadata_schema_uri = 3
Required. Points to a YAML file stored on Google Cloud Storage describing additional information about the Dataset. The schema is defined as an OpenAPI 3.0.2 Schema Object. The schema files that can be used here are found in gs://google-cloud-aiplatform/schema/dataset/metadata/.
optional protobuf.Value metadata = 8
Required. Additional information about the Dataset.
int64 data_item_count = 10
Output only. The number of DataItems in this Dataset. Only apply for non-structured Dataset.
optional protobuf.Timestamp create_time = 4
Output only. Timestamp when this Dataset was created.
optional protobuf.Timestamp update_time = 5
Output only. Timestamp when this Dataset was last updated.
string etag = 6
Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 7
The labels with user-defined metadata to organize your Datasets. 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. No more than 64 user labels can be associated with one Dataset (System labels are excluded). See https://goo.gl/xmQnxf for more information and examples of labels. System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable. Following system labels exist for each Dataset: * "aiplatform.googleapis.com/dataset_metadata_schema": output only, its value is the [metadata_schema's][google.cloud.aiplatform.v1.Dataset.metadata_schema_uri] title.
repeated SavedQuery saved_queries = 9
All SavedQueries belong to the Dataset will be returned in List/Get Dataset response. The annotation_specs field will not be populated except for UI cases which will only use [annotation_spec_count][google.cloud.aiplatform.v1.SavedQuery.annotation_spec_count]. In CreateDataset request, a SavedQuery is created together if this field is set, up to one SavedQuery can be set in CreateDatasetRequest. The SavedQuery should not contain any AnnotationSpec.
optional EncryptionSpec encryption_spec = 11
Customer-managed encryption key spec for a Dataset. If set, this Dataset and all sub-resources of this Dataset will be secured by this key.
string metadata_artifact = 17
Output only. The resource name of the Artifact that was created in MetadataStore when creating the Dataset. The Artifact resource name pattern is `projects/{project}/locations/{location}/metadataStores/{metadata_store}/artifacts/{artifact}`.
string model_reference = 18
Optional. Reference to the public base model last used by the dataset. Only set for prompt datasets.
bool satisfies_pzs = 19
Output only. Reserved for future use.
bool satisfies_pzi = 20
Output only. Reserved for future use.

Describes the dataset version.

Used as response type in: DatasetService.GetDatasetVersion, DatasetService.UpdateDatasetVersion

Used as field type in: CreateDatasetVersionRequest, ListDatasetVersionsResponse, UpdateDatasetVersionRequest

string name = 1
Output only. Identifier. The resource name of the DatasetVersion.
optional protobuf.Timestamp create_time = 2
Output only. Timestamp when this DatasetVersion was created.
optional protobuf.Timestamp update_time = 6
Output only. Timestamp when this DatasetVersion was last updated.
string etag = 3
Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
string big_query_dataset_name = 4
Output only. Name of the associated BigQuery dataset.
string display_name = 7
The user-defined name of the DatasetVersion. The name can be up to 128 characters long and can consist of any UTF-8 characters.
optional protobuf.Value metadata = 8
Required. Output only. Additional information about the DatasetVersion.
string model_reference = 9
Output only. Reference to the public base model last used by the dataset version. Only set for prompt dataset versions.
bool satisfies_pzs = 10
Output only. Reserved for future use.
bool satisfies_pzi = 11
Output only. Reserved for future use.

A description of resources that are dedicated to a DeployedModel, and that need a higher degree of manual configuration.

Used in: DeployedIndex, DeployedModel, DeploymentResourcePool, PublisherModel.CallToAction.Deploy

optional MachineSpec machine_spec = 1
Required. Immutable. The specification of a single machine used by the prediction.
int32 min_replica_count = 2
Required. Immutable. The minimum number of machine replicas this DeployedModel will be always deployed on. This value must be greater than or equal to 1. If traffic against the DeployedModel increases, it may dynamically be deployed onto more replicas, and as traffic decreases, some of these extra replicas may be freed.
int32 max_replica_count = 3
Immutable. The maximum number of replicas this DeployedModel may be deployed on when the traffic against it increases. If the requested value is too large, the deployment will error, but if deployment succeeds then the ability to scale the model to that many replicas is guaranteed (barring service outages). If traffic against the DeployedModel increases beyond what its replicas at maximum may handle, a portion of the traffic will be dropped. If this value is not provided, will use [min_replica_count][google.cloud.aiplatform.v1.DedicatedResources.min_replica_count] as the default value. The value of this field impacts the charge against Vertex CPU and GPU quotas. Specifically, you will be charged for (max_replica_count * number of cores in the selected machine type) and (max_replica_count * number of GPUs per replica in the selected machine type).
int32 required_replica_count = 9
Optional. Number of required available replicas for the deployment to succeed. This field is only needed when partial model deployment/mutation is desired. If set, the model deploy/mutate operation will succeed once available_replica_count reaches required_replica_count, and the rest of the replicas will be retried. If not set, the default required_replica_count will be min_replica_count.
repeated AutoscalingMetricSpec autoscaling_metric_specs = 4
Immutable. The metric specifications that overrides a resource utilization metric (CPU utilization, accelerator's duty cycle, and so on) target value (default to 60 if not set). At most one entry is allowed per metric. If [machine_spec.accelerator_count][google.cloud.aiplatform.v1.MachineSpec.accelerator_count] is above 0, the autoscaling will be based on both CPU utilization and accelerator's duty cycle metrics and scale up when either metrics exceeds its target value while scale down if both metrics are under their target value. The default target value is 60 for both metrics. If [machine_spec.accelerator_count][google.cloud.aiplatform.v1.MachineSpec.accelerator_count] is 0, the autoscaling will be based on CPU utilization metric only with default target value 60 if not explicitly set. For example, in the case of Online Prediction, if you want to override target CPU utilization to 80, you should set [autoscaling_metric_specs.metric_name][google.cloud.aiplatform.v1.AutoscalingMetricSpec.metric_name] to `aiplatform.googleapis.com/prediction/online/cpu/utilization` and [autoscaling_metric_specs.target][google.cloud.aiplatform.v1.AutoscalingMetricSpec.target] to `80`.
bool spot = 5
Optional. If true, schedule the deployment workload on [spot VMs](https://cloud.google.com/kubernetes-engine/docs/concepts/spot-vms).

Request message for [FeaturestoreService.DeleteFeature][google.cloud.aiplatform.v1.FeaturestoreService.DeleteFeature]. Request message for [FeatureRegistryService.DeleteFeature][google.cloud.aiplatform.v1.FeatureRegistryService.DeleteFeature].

Used as request type in: FeatureRegistryService.DeleteFeature, FeaturestoreService.DeleteFeature

string name = 1
Required. The name of the Features to be deleted. Format: `projects/{project}/locations/{location}/featurestores/{featurestore}/entityTypes/{entity_type}/features/{feature}` `projects/{project}/locations/{location}/featureGroups/{feature_group}/features/{feature}`

Details of operations that delete Feature values.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for Featurestore delete Features values.

Message to select entity. If an entity id is selected, all the feature values corresponding to the entity id will be deleted, including the entityId.

Used in: DeleteFeatureValuesRequest

optional EntityIdSelector entity_id_selector = 1
Required. Selectors choosing feature values of which entity id to be deleted from the EntityType.

Message to select time range and feature. Values of the selected feature generated within an inclusive time range will be deleted. Using this option permanently deletes the feature values from the specified feature IDs within the specified time range. This might include data from the online storage. If you want to retain any deleted historical data in the online storage, you must re-ingest it.

Used in: DeleteFeatureValuesRequest

optional type.Interval time_range = 1
Required. Select feature generated within a half-inclusive time range. The time range is lower inclusive and upper exclusive.
optional FeatureSelector feature_selector = 2
Required. Selectors choosing which feature values to be deleted from the EntityType.
bool skip_online_storage_delete = 3
If set, data will not be deleted from online storage. When time range is older than the data in online storage, setting this to be true will make the deletion have no impact on online serving.

Response message for [FeaturestoreService.DeleteFeatureValues][google.cloud.aiplatform.v1.FeaturestoreService.DeleteFeatureValues].

oneof response
Response based on which delete option is specified in the request
- DeleteFeatureValuesResponse.SelectEntity select_entity = 1
  Response for request specifying the entities to delete
- DeleteFeatureValuesResponse.SelectTimeRangeAndFeature select_time_range_and_feature = 2
  Response for request specifying time range and feature

Response message if the request uses the SelectEntity option.

Used in: DeleteFeatureValuesResponse

int64 offline_storage_deleted_entity_row_count = 1
The count of deleted entity rows in the offline storage. Each row corresponds to the combination of an entity ID and a timestamp. One entity ID can have multiple rows in the offline storage.
int64 online_storage_deleted_entity_count = 2
The count of deleted entities in the online storage. Each entity ID corresponds to one entity.

Response message if the request uses the SelectTimeRangeAndFeature option.

Used in: DeleteFeatureValuesResponse

int64 impacted_feature_count = 1
The count of the features or columns impacted. This is the same as the feature count in the request.
int64 offline_storage_modified_entity_row_count = 2
The count of modified entity rows in the offline storage. Each row corresponds to the combination of an entity ID and a timestamp. One entity ID can have multiple rows in the offline storage. Within each row, only the features specified in the request are deleted.
int64 online_storage_modified_entity_count = 3
The count of modified entities in the online storage. Each entity ID corresponds to one entity. Within each entity, only the features specified in the request are deleted.

Details of operations that perform [MetadataService.DeleteMetadataStore][google.cloud.aiplatform.v1.MetadataService.DeleteMetadataStore].

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for deleting a MetadataStore.

Details of operations that perform deletes of any entities.

optional GenericOperationMetadata generic_metadata = 1
The common part of the operation metadata.

Runtime operation information for [IndexEndpointService.DeployIndex][google.cloud.aiplatform.v1.IndexEndpointService.DeployIndex].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.
string deployed_index_id = 2
The unique index id specified by user

Response message for [IndexEndpointService.DeployIndex][google.cloud.aiplatform.v1.IndexEndpointService.DeployIndex].

optional DeployedIndex deployed_index = 1
The DeployedIndex that had been deployed in the IndexEndpoint.

Runtime operation information for [EndpointService.DeployModel][google.cloud.aiplatform.v1.EndpointService.DeployModel].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Response message for [EndpointService.DeployModel][google.cloud.aiplatform.v1.EndpointService.DeployModel].

optional DeployedModel deployed_model = 1
The DeployedModel that had been deployed in the Endpoint.

A deployment of an Index. IndexEndpoints contain one or more DeployedIndexes.

Used in: DeployIndexRequest, DeployIndexResponse, IndexEndpoint, MutateDeployedIndexRequest, MutateDeployedIndexResponse

string id = 1
Required. The user specified ID of the DeployedIndex. The ID can be up to 128 characters long and must start with a letter and only contain letters, numbers, and underscores. The ID must be unique within the project it is created in.
string index = 2
Required. The name of the Index this is the deployment of. We may refer to this Index as the DeployedIndex's "original" Index.
string display_name = 3
The display name of the DeployedIndex. If not provided upon creation, the Index's display_name is used.
optional protobuf.Timestamp create_time = 4
Output only. Timestamp when the DeployedIndex was created.
optional IndexPrivateEndpoints private_endpoints = 5
Output only. Provides paths for users to send requests directly to the deployed index services running on Cloud via private services access. This field is populated if [network][google.cloud.aiplatform.v1.IndexEndpoint.network] is configured.
optional protobuf.Timestamp index_sync_time = 6
Output only. The DeployedIndex may depend on various data on its original Index. Additionally when certain changes to the original Index are being done (e.g. when what the Index contains is being changed) the DeployedIndex may be asynchronously updated in the background to reflect these changes. If this timestamp's value is at least the [Index.update_time][google.cloud.aiplatform.v1.Index.update_time] of the original Index, it means that this DeployedIndex and the original Index are in sync. If this timestamp is older, then to see which updates this DeployedIndex already contains (and which it does not), one must [list][google.longrunning.Operations.ListOperations] the operations that are running on the original Index. Only the successfully completed Operations with [update_time][google.cloud.aiplatform.v1.GenericOperationMetadata.update_time] equal or before this sync time are contained in this DeployedIndex.
optional AutomaticResources automatic_resources = 7
Optional. A description of resources that the DeployedIndex uses, which to large degree are decided by Vertex AI, and optionally allows only a modest additional configuration. If min_replica_count is not set, the default value is 2 (we don't provide SLA when min_replica_count=1). If max_replica_count is not set, the default value is min_replica_count. The max allowed replica count is 1000.
optional DedicatedResources dedicated_resources = 16
Optional. A description of resources that are dedicated to the DeployedIndex, and that need a higher degree of manual configuration. The field min_replica_count must be set to a value strictly greater than 0, or else validation will fail. We don't provide SLA when min_replica_count=1. If max_replica_count is not set, the default value is min_replica_count. The max allowed replica count is 1000. Available machine types for SMALL shard: e2-standard-2 and all machine types available for MEDIUM and LARGE shard. Available machine types for MEDIUM shard: e2-standard-16 and all machine types available for LARGE shard. Available machine types for LARGE shard: e2-highmem-16, n2d-standard-32. n1-standard-16 and n1-standard-32 are still available, but we recommend e2-standard-16 and e2-highmem-16 for cost efficiency.
bool enable_access_logging = 8
Optional. If true, private endpoint's access logs are sent to Cloud Logging. These logs are like standard server access logs, containing information like timestamp and latency for each MatchRequest. Note that logs may incur a cost, especially if the deployed index receives a high queries per second rate (QPS). Estimate your costs before enabling this option.
optional DeployedIndexAuthConfig deployed_index_auth_config = 9
Optional. If set, the authentication is enabled for the private endpoint.
repeated string reserved_ip_ranges = 10
Optional. A list of reserved ip ranges under the VPC network that can be used for this DeployedIndex. If set, we will deploy the index within the provided ip ranges. Otherwise, the index might be deployed to any ip ranges under the provided VPC network. The value should be the name of the address (https://cloud.google.com/compute/docs/reference/rest/v1/addresses) Example: ['vertex-ai-ip-range']. For more information about subnets and network IP ranges, please see https://cloud.google.com/vpc/docs/subnets#manually_created_subnet_ip_ranges.
string deployment_group = 11
Optional. The deployment group can be no longer than 64 characters (eg: 'test', 'prod'). If not set, we will use the 'default' deployment group. Creating `deployment_groups` with `reserved_ip_ranges` is a recommended practice when the peered network has multiple peering ranges. This creates your deployments from predictable IP spaces for easier traffic administration. Also, one deployment_group (except 'default') can only be used with the same reserved_ip_ranges which means if the deployment_group has been used with reserved_ip_ranges: [a, b, c], using it with [a, b] or [d, e] is disallowed. Note: we only support up to 5 deployment groups(not including 'default').
repeated PSCAutomationConfig psc_automation_configs = 19
Optional. If set for PSC deployed index, PSC connection will be automatically created after deployment is done and the endpoint information is populated in private_endpoints.psc_automated_endpoints.

Used to set up the auth on the DeployedIndex's private endpoint.

Used in: DeployedIndex

optional DeployedIndexAuthConfig.AuthProvider auth_provider = 1
Defines the authentication provider that the DeployedIndex uses.

Configuration for an authentication provider, including support for [JSON Web Token (JWT)](https://tools.ietf.org/html/draft-ietf-oauth-json-web-token-32).

Used in: DeployedIndexAuthConfig

repeated string audiences = 1
The list of JWT [audiences](https://tools.ietf.org/html/draft-ietf-oauth-json-web-token-32#section-4.1.3). that are allowed to access. A JWT containing any of these audiences will be accepted.
repeated string allowed_issuers = 2
A list of allowed JWT issuers. Each entry must be a valid Google service account, in the following format: `service-account-name@project-id.iam.gserviceaccount.com`

Points to a DeployedIndex.

Used in: Index

string index_endpoint = 1
Immutable. A resource name of the IndexEndpoint.
string deployed_index_id = 2
Immutable. The ID of the DeployedIndex in the above IndexEndpoint.
string display_name = 3
Output only. The display name of the DeployedIndex.

A deployment of a Model. Endpoints contain one or more DeployedModels.

Used in: DeployModelRequest, DeployModelResponse, Endpoint, MutateDeployedModelRequest, MutateDeployedModelResponse, QueryDeployedModelsResponse

oneof prediction_resources
The prediction (for example, the machine) resources that the DeployedModel uses. The user is billed for the resources (at least their minimal amount) even if the DeployedModel receives no traffic. Not all Models support all resources types. See [Model.supported_deployment_resources_types][google.cloud.aiplatform.v1.Model.supported_deployment_resources_types]. Required except for Large Model Deploy use cases.
- DedicatedResources dedicated_resources = 7
  A description of resources that are dedicated to the DeployedModel, and that need a higher degree of manual configuration.
- AutomaticResources automatic_resources = 8
  A description of resources that to large degree are decided by Vertex AI, and require only a modest additional configuration.
- string shared_resources = 17
  The resource name of the shared DeploymentResourcePool to deploy on. Format: `projects/{project}/locations/{location}/deploymentResourcePools/{deployment_resource_pool}`
string id = 1
Immutable. The ID of the DeployedModel. If not provided upon deployment, Vertex AI will generate a value for this ID. This value should be 1-10 characters, and valid characters are `/[0-9]/`.
string model = 2
Required. The resource name of the Model that this is the deployment of. Note that the Model may be in a different location than the DeployedModel's Endpoint. The resource name may contain version id or version alias to specify the version. Example: `projects/{project}/locations/{location}/models/{model}@2` or `projects/{project}/locations/{location}/models/{model}@golden` if no version is specified, the default version will be deployed.
string model_version_id = 18
Output only. The version ID of the model that is deployed.
string display_name = 3
The display name of the DeployedModel. If not provided upon creation, the Model's display_name is used.
optional protobuf.Timestamp create_time = 6
Output only. Timestamp when the DeployedModel was created.
optional ExplanationSpec explanation_spec = 9
Explanation configuration for this DeployedModel. When deploying a Model using [EndpointService.DeployModel][google.cloud.aiplatform.v1.EndpointService.DeployModel], this value overrides the value of [Model.explanation_spec][google.cloud.aiplatform.v1.Model.explanation_spec]. All fields of [explanation_spec][google.cloud.aiplatform.v1.DeployedModel.explanation_spec] are optional in the request. If a field of [explanation_spec][google.cloud.aiplatform.v1.DeployedModel.explanation_spec] is not populated, the value of the same field of [Model.explanation_spec][google.cloud.aiplatform.v1.Model.explanation_spec] is inherited. If the corresponding [Model.explanation_spec][google.cloud.aiplatform.v1.Model.explanation_spec] is not populated, all fields of the [explanation_spec][google.cloud.aiplatform.v1.DeployedModel.explanation_spec] will be used for the explanation configuration.
bool disable_explanations = 19
If true, deploy the model without explainable feature, regardless the existence of [Model.explanation_spec][google.cloud.aiplatform.v1.Model.explanation_spec] or [explanation_spec][google.cloud.aiplatform.v1.DeployedModel.explanation_spec].
string service_account = 11
The service account that the DeployedModel's container runs as. Specify the email address of the service account. If this service account is not specified, the container runs as a service account that doesn't have access to the resource project. Users deploying the Model must have the `iam.serviceAccounts.actAs` permission on this service account.
bool disable_container_logging = 15
For custom-trained Models and AutoML Tabular Models, the container of the DeployedModel instances will send `stderr` and `stdout` streams to Cloud Logging by default. Please note that the logs incur cost, which are subject to [Cloud Logging pricing](https://cloud.google.com/logging/pricing). User can disable container logging by setting this flag to true.
bool enable_access_logging = 13
If true, online prediction access logs are sent to Cloud Logging. These logs are like standard server access logs, containing information like timestamp and latency for each prediction request. Note that logs may incur a cost, especially if your project receives prediction requests at a high queries per second rate (QPS). Estimate your costs before enabling this option.
optional PrivateEndpoints private_endpoints = 14
Output only. Provide paths for users to send predict/explain/health requests directly to the deployed model services running on Cloud via private services access. This field is populated if [network][google.cloud.aiplatform.v1.Endpoint.network] is configured.
optional FasterDeploymentConfig faster_deployment_config = 23
Configuration for faster model deployment.
optional DeployedModel.Status status = 26
Output only. Runtime status of the deployed model.
map<string, string> system_labels = 28
System labels to apply to Model Garden deployments. System labels are managed by Google for internal use only.
optional SpeculativeDecodingSpec speculative_decoding_spec = 30
Optional. Spec for configuring speculative decoding.

Runtime status of the deployed model.

Used in: DeployedModel

string message = 1
Output only. The latest deployed model's status message (if any).
optional protobuf.Timestamp last_update_time = 2
Output only. The time at which the status was last updated.
int32 available_replica_count = 3
Output only. The number of available replicas of the deployed model.

Points to a DeployedModel.

Used in: Model, QueryDeployedModelsResponse

string endpoint = 1
Immutable. A resource name of an Endpoint.
string deployed_model_id = 2
Immutable. An ID of a DeployedModel in the above Endpoint.

A description of resources that can be shared by multiple DeployedModels, whose underlying specification consists of a DedicatedResources.

Used as response type in: DeploymentResourcePoolService.GetDeploymentResourcePool

Used as field type in: CreateDeploymentResourcePoolRequest, ListDeploymentResourcePoolsResponse, UpdateDeploymentResourcePoolRequest

string name = 1
Immutable. The resource name of the DeploymentResourcePool. Format: `projects/{project}/locations/{location}/deploymentResourcePools/{deployment_resource_pool}`
optional DedicatedResources dedicated_resources = 2
Required. The underlying DedicatedResources that the DeploymentResourcePool uses.
optional EncryptionSpec encryption_spec = 5
Customer-managed encryption key spec for a DeploymentResourcePool. If set, this DeploymentResourcePool will be secured by this key. Endpoints and the DeploymentResourcePool they deploy in need to have the same EncryptionSpec.
string service_account = 6
The service account that the DeploymentResourcePool's container(s) run as. Specify the email address of the service account. If this service account is not specified, the container(s) run as a service account that doesn't have access to the resource project. Users deploying the Models to this DeploymentResourcePool must have the `iam.serviceAccounts.actAs` permission on this service account.
bool disable_container_logging = 7
If the DeploymentResourcePool is deployed with custom-trained Models or AutoML Tabular Models, the container(s) of the DeploymentResourcePool will send `stderr` and `stdout` streams to Cloud Logging by default. Please note that the logs incur cost, which are subject to [Cloud Logging pricing](https://cloud.google.com/logging/pricing). User can disable container logging by setting this flag to true.
optional protobuf.Timestamp create_time = 4
Output only. Timestamp when this DeploymentResourcePool was created.
bool satisfies_pzs = 8
Output only. Reserved for future use.
bool satisfies_pzi = 9
Output only. Reserved for future use.

Used in: BatchReadFeatureValuesRequest.EntityTypeSpec, ExportFeatureValuesRequest

string feature_id = 1
Required. The ID of the Feature to apply the setting to.
string destination_field = 2
Specify the field name in the export destination. If not specified, Feature ID is used.

The input content is encapsulated and uploaded in the request.

Used in: RagFile

(message has no fields)

Represents the spec of disk options.

Used in: ResourcePool, WorkerPoolSpec

string boot_disk_type = 1
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).
int32 boot_disk_size_gb = 2
Size in GB of the boot disk (default is 100GB).

A list of double values.

Used in: FeatureValue

repeated double values = 1
A list of double values.

Describes the options to customize dynamic retrieval.

Used in: GoogleSearchRetrieval

DynamicRetrievalConfig.Mode mode = 1
The mode of the predictor to be used in dynamic retrieval.
optional float dynamic_threshold = 2
Optional. The threshold to be used in dynamic retrieval. If not set, a system default value is used.

The mode of the predictor to be used in dynamic retrieval.

Used in: DynamicRetrievalConfig

MODE_UNSPECIFIED = 0
Always trigger retrieval.
MODE_DYNAMIC = 1
Run retrieval only when system decides it is necessary.

Represents a customer-managed encryption key spec that can be applied to a top-level resource.

Used in: BatchPredictionJob, CachedContent, CopyModelRequest, CustomJob, DataLabelingJob, Dataset, DeploymentResourcePool, Endpoint, FeatureOnlineStore, Featurestore, HyperparameterTuningJob, Index, IndexEndpoint, MetadataStore, Model, ModelDeploymentMonitoringJob, NasJob, NotebookExecutionJob, NotebookRuntime, NotebookRuntimeTemplate, PersistentResource, PipelineJob, Tensorboard, TrainingPipeline, TuningJob

string kms_key_name = 1
Required. The Cloud KMS resource identifier of the customer managed encryption key used to protect a resource. Has the form: `projects/my-project/locations/my-region/keyRings/my-kr/cryptoKeys/my-key`. The key needs to be in the same region as where the compute resource is created.

Models are deployed into it, and afterwards Endpoint is called to obtain predictions and explanations.

Used as response type in: EndpointService.GetEndpoint, EndpointService.UpdateEndpoint

Used as field type in: CreateEndpointRequest, ListEndpointsResponse, UpdateEndpointLongRunningRequest, UpdateEndpointRequest

string name = 1
Output only. The resource name of the Endpoint.
string display_name = 2
Required. The display name of the Endpoint. The name can be up to 128 characters long and can consist of any UTF-8 characters.
string description = 3
The description of the Endpoint.
repeated DeployedModel deployed_models = 4
Output only. The models deployed in this Endpoint. To add or remove DeployedModels use [EndpointService.DeployModel][google.cloud.aiplatform.v1.EndpointService.DeployModel] and [EndpointService.UndeployModel][google.cloud.aiplatform.v1.EndpointService.UndeployModel] respectively.
map<string, int32> traffic_split = 5
A map from a DeployedModel's ID to the percentage of this Endpoint's traffic that should be forwarded to that DeployedModel. If a DeployedModel's ID is not listed in this map, then it receives no traffic. The traffic percentage values must add up to 100, or map must be empty if the Endpoint is to not accept any traffic at a moment.
string etag = 6
Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 7
The labels with user-defined metadata to organize your Endpoints. 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.
optional protobuf.Timestamp create_time = 8
Output only. Timestamp when this Endpoint was created.
optional protobuf.Timestamp update_time = 9
Output only. Timestamp when this Endpoint was last updated.
optional EncryptionSpec encryption_spec = 10
Customer-managed encryption key spec for an Endpoint. If set, this Endpoint and all sub-resources of this Endpoint will be secured by this key.
string network = 13
Optional. The full name of the Google Compute Engine [network](https://cloud.google.com//compute/docs/networks-and-firewalls#networks) to which the Endpoint should be peered. Private services access must already be configured for the network. If left unspecified, the Endpoint is not peered with any network. Only one of the fields, [network][google.cloud.aiplatform.v1.Endpoint.network] or [enable_private_service_connect][google.cloud.aiplatform.v1.Endpoint.enable_private_service_connect], can be set. [Format](https://cloud.google.com/compute/docs/reference/rest/v1/networks/insert): `projects/{project}/global/networks/{network}`. Where `{project}` is a project number, as in `12345`, and `{network}` is network name.
bool enable_private_service_connect = 17
Deprecated: If true, expose the Endpoint via private service connect. Only one of the fields, [network][google.cloud.aiplatform.v1.Endpoint.network] or [enable_private_service_connect][google.cloud.aiplatform.v1.Endpoint.enable_private_service_connect], can be set.
optional PrivateServiceConnectConfig private_service_connect_config = 21
Optional. Configuration for private service connect. [network][google.cloud.aiplatform.v1.Endpoint.network] and [private_service_connect_config][google.cloud.aiplatform.v1.Endpoint.private_service_connect_config] are mutually exclusive.
string model_deployment_monitoring_job = 14
Output only. Resource name of the Model Monitoring job associated with this Endpoint if monitoring is enabled by [JobService.CreateModelDeploymentMonitoringJob][google.cloud.aiplatform.v1.JobService.CreateModelDeploymentMonitoringJob]. Format: `projects/{project}/locations/{location}/modelDeploymentMonitoringJobs/{model_deployment_monitoring_job}`
optional PredictRequestResponseLoggingConfig predict_request_response_logging_config = 18
Configures the request-response logging for online prediction.
bool dedicated_endpoint_enabled = 24
If true, the endpoint will be exposed through a dedicated DNS [Endpoint.dedicated_endpoint_dns]. Your request to the dedicated DNS will be isolated from other users' traffic and will have better performance and reliability. Note: Once you enabled dedicated endpoint, you won't be able to send request to the shared DNS {region}-aiplatform.googleapis.com. The limitation will be removed soon.
string dedicated_endpoint_dns = 25
Output only. DNS of the dedicated endpoint. Will only be populated if dedicated_endpoint_enabled is true. Format: `https://{endpoint_id}.{region}-{project_number}.prediction.vertexai.goog`.
optional ClientConnectionConfig client_connection_config = 23
Configurations that are applied to the endpoint for online prediction.
bool satisfies_pzs = 27
Output only. Reserved for future use.
bool satisfies_pzi = 28
Output only. Reserved for future use.

Tool to search public web data, powered by Vertex AI Search and Sec4 compliance.

Used in: Tool

(message has no fields)

Selector for entityId. Getting ids from the given source.

Used in: DeleteFeatureValuesRequest.SelectEntity

oneof EntityIdsSource
Details about the source data, including the location of the storage and the format.
- CsvSource csv_source = 3
  Source of Csv
string entity_id_field = 5
Source column that holds entity IDs. If not provided, entity IDs are extracted from the column named entity_id.

An entity type is a type of object in a system that needs to be modeled and have stored information about. For example, driver is an entity type, and driver0 is an instance of an entity type driver.

Used as response type in: FeaturestoreService.GetEntityType, FeaturestoreService.UpdateEntityType

Used as field type in: CreateEntityTypeRequest, ListEntityTypesResponse, UpdateEntityTypeRequest

string name = 1
Immutable. Name of the EntityType. Format: `projects/{project}/locations/{location}/featurestores/{featurestore}/entityTypes/{entity_type}` The last part entity_type is assigned by the client. The entity_type can be up to 64 characters long and can consist only of ASCII Latin letters A-Z and a-z and underscore(_), and ASCII digits 0-9 starting with a letter. The value will be unique given a featurestore.
string description = 2
Optional. Description of the EntityType.
optional protobuf.Timestamp create_time = 3
Output only. Timestamp when this EntityType was created.
optional protobuf.Timestamp update_time = 4
Output only. Timestamp when this EntityType was most recently updated.
map<string, string> labels = 6
Optional. The labels with user-defined metadata to organize your EntityTypes. 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 on and examples of labels. No more than 64 user labels can be associated with one EntityType (System labels are excluded)." System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
string etag = 7
Optional. Used to perform a consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
optional FeaturestoreMonitoringConfig monitoring_config = 8
Optional. The default monitoring configuration for all Features with value type ([Feature.ValueType][google.cloud.aiplatform.v1.Feature.ValueType]) BOOL, STRING, DOUBLE or INT64 under this EntityType. If this is populated with [FeaturestoreMonitoringConfig.monitoring_interval] specified, snapshot analysis monitoring is enabled. Otherwise, snapshot analysis monitoring is disabled.
int32 offline_storage_ttl_days = 10
Optional. Config for data retention policy in offline storage. TTL in days for feature values that will be stored in offline storage. The Feature Store offline storage periodically removes obsolete feature values older than `offline_storage_ttl_days` since the feature generation time. If unset (or explicitly set to 0), default to 4000 days TTL.
bool satisfies_pzs = 11
Output only. Reserved for future use.
bool satisfies_pzi = 12
Output only. Reserved for future use.

Represents an environment variable present in a Container or Python Module.

Used in: ContainerSpec, ModelContainerSpec, NotebookSoftwareConfig, PythonPackageSpec, ReasoningEngineSpec.DeploymentSpec

string name = 1
Required. Name of the environment variable. Must be a valid C identifier.
string value = 2
Required. Variables that reference a $(VAR_NAME) are expanded using the previous defined environment variables in the container and any service environment variables. If a variable cannot be resolved, the reference in the input string will be unchanged. The $(VAR_NAME) syntax can be escaped with a double $$, ie: $$(VAR_NAME). Escaped references will never be expanded, regardless of whether the variable exists or not.

Model error analysis for each annotation.

Used in: EvaluatedAnnotation

repeated ErrorAnalysisAnnotation.AttributedItem attributed_items = 1
Attributed items for a given annotation, typically representing neighbors from the training sets constrained by the query type.
ErrorAnalysisAnnotation.QueryType query_type = 2
The query type used for finding the attributed items.
double outlier_score = 3
The outlier score of this annotated item. Usually defined as the min of all distances from attributed items.
double outlier_threshold = 4
The threshold used to determine if this annotation is an outlier or not.

Attributed items for a given annotation, typically representing neighbors from the training sets constrained by the query type.

Used in: ErrorAnalysisAnnotation

string annotation_resource_name = 1
The unique ID for each annotation. Used by FE to allocate the annotation in DB.
double distance = 2
The distance of this item to the annotation.

The query type used for finding the attributed items.

Used in: ErrorAnalysisAnnotation

QUERY_TYPE_UNSPECIFIED = 0
Unspecified query type for model error analysis.
ALL_SIMILAR = 1
Query similar samples across all classes in the dataset.
SAME_CLASS_SIMILAR = 2
Query similar samples from the same class of the input sample.
SAME_CLASS_DISSIMILAR = 3
Query dissimilar samples from the same class of the input sample.

True positive, false positive, or false negative. EvaluatedAnnotation is only available under ModelEvaluationSlice with slice of `annotationSpec` dimension.

Used in: BatchImportEvaluatedAnnotationsRequest

EvaluatedAnnotation.EvaluatedAnnotationType type = 1
Output only. Type of the EvaluatedAnnotation.
repeated protobuf.Value predictions = 2
Output only. The model predicted annotations. For true positive, there is one and only one prediction, which matches the only one ground truth annotation in [ground_truths][google.cloud.aiplatform.v1.EvaluatedAnnotation.ground_truths]. For false positive, there is one and only one prediction, which doesn't match any ground truth annotation of the corresponding [data_item_view_id][google.cloud.aiplatform.v1.EvaluatedAnnotation.evaluated_data_item_view_id]. For false negative, there are zero or more predictions which are similar to the only ground truth annotation in [ground_truths][google.cloud.aiplatform.v1.EvaluatedAnnotation.ground_truths] but not enough for a match. The schema of the prediction is stored in [ModelEvaluation.annotation_schema_uri][google.cloud.aiplatform.v1.ModelEvaluation.annotation_schema_uri]
repeated protobuf.Value ground_truths = 3
Output only. The ground truth Annotations, i.e. the Annotations that exist in the test data the Model is evaluated on. For true positive, there is one and only one ground truth annotation, which matches the only prediction in [predictions][google.cloud.aiplatform.v1.EvaluatedAnnotation.predictions]. For false positive, there are zero or more ground truth annotations that are similar to the only prediction in [predictions][google.cloud.aiplatform.v1.EvaluatedAnnotation.predictions], but not enough for a match. For false negative, there is one and only one ground truth annotation, which doesn't match any predictions created by the model. The schema of the ground truth is stored in [ModelEvaluation.annotation_schema_uri][google.cloud.aiplatform.v1.ModelEvaluation.annotation_schema_uri]
optional protobuf.Value data_item_payload = 5
Output only. The data item payload that the Model predicted this EvaluatedAnnotation on.
string evaluated_data_item_view_id = 6
Output only. ID of the EvaluatedDataItemView under the same ancestor ModelEvaluation. The EvaluatedDataItemView consists of all ground truths and predictions on [data_item_payload][google.cloud.aiplatform.v1.EvaluatedAnnotation.data_item_payload].
repeated EvaluatedAnnotationExplanation explanations = 8
Explanations of [predictions][google.cloud.aiplatform.v1.EvaluatedAnnotation.predictions]. Each element of the explanations indicates the explanation for one explanation Method. The attributions list in the [EvaluatedAnnotationExplanation.explanation][google.cloud.aiplatform.v1.EvaluatedAnnotationExplanation.explanation] object corresponds to the [predictions][google.cloud.aiplatform.v1.EvaluatedAnnotation.predictions] list. For example, the second element in the attributions list explains the second element in the predictions list.
repeated ErrorAnalysisAnnotation error_analysis_annotations = 9
Annotations of model error analysis results.

Describes the type of the EvaluatedAnnotation. The type is determined

Used in: EvaluatedAnnotation

EVALUATED_ANNOTATION_TYPE_UNSPECIFIED = 0
Invalid value.
TRUE_POSITIVE = 1
The EvaluatedAnnotation is a true positive. It has a prediction created by the Model and a ground truth Annotation which the prediction matches.
FALSE_POSITIVE = 2
The EvaluatedAnnotation is false positive. It has a prediction created by the Model which does not match any ground truth annotation.
FALSE_NEGATIVE = 3
The EvaluatedAnnotation is false negative. It has a ground truth annotation which is not matched by any of the model created predictions.

Explanation result of the prediction produced by the Model.

Used in: EvaluatedAnnotation

string explanation_type = 1
Explanation type. For AutoML Image Classification models, possible values are: * `image-integrated-gradients` * `image-xrai`
optional Explanation explanation = 2
Explanation attribution response details.

An edge describing the relationship between an Artifact and an Execution in a lineage graph.

Used in: AddExecutionEventsRequest, LineageSubgraph

string artifact = 1
Required. The relative resource name of the Artifact in the Event.
string execution = 2
Output only. The relative resource name of the Execution in the Event.
optional protobuf.Timestamp event_time = 3
Output only. Time the Event occurred.
Event.Type type = 4
Required. The type of the Event.
map<string, string> labels = 5
The labels with user-defined metadata to annotate Events. 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. No more than 64 user labels can be associated with one Event (System labels are excluded). See https://goo.gl/xmQnxf for more information and examples of labels. System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.

Describes whether an Event's Artifact is the Execution's input or output.

Used in: Event

TYPE_UNSPECIFIED = 0
Unspecified whether input or output of the Execution.
INPUT = 1
An input of the Execution.
OUTPUT = 2
An output of the Execution.

Input for exact match metric.

optional ExactMatchSpec metric_spec = 1
Required. Spec for exact match metric.
repeated ExactMatchInstance instances = 2
Required. Repeated exact match instances.

Spec for exact match instance.

Used in: ExactMatchInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Required. Ground truth used to compare against the prediction.

Exact match metric value for an instance.

Used in: ExactMatchResults

optional float score = 1
Output only. Exact match score.

Results for exact match metric.

Used in: BatchReadFeatureValuesRequest.EntityTypeSpec, DeleteFeatureValuesRequest.SelectTimeRangeAndFeature, ExportFeatureValuesRequest, ReadFeatureValuesRequest, StreamingReadFeatureValuesRequest

repeated ExactMatchMetricValue exact_match_metric_values = 1
Output only. Exact match metric values.

Spec for exact match metric - returns 1 if prediction and reference exactly matches, otherwise 0.

Used in: ExactMatchInput

(message has no fields)

Example-based explainability that returns the nearest neighbors from the provided dataset.

Used in: ExplanationParameters, UpdateExplanationDatasetRequest

oneof source
- Examples.ExampleGcsSource example_gcs_source = 5
  The Cloud Storage input instances.
oneof config
- protobuf.Value nearest_neighbor_search_config = 2
  The full configuration for the generated index, the semantics are the same as [metadata][google.cloud.aiplatform.v1.Index.metadata] and should match [NearestNeighborSearchConfig](https://cloud.google.com/vertex-ai/docs/explainable-ai/configuring-explanations-example-based#nearest-neighbor-search-config).
- Presets presets = 4
  Simplified preset configuration, which automatically sets configuration values based on the desired query speed-precision trade-off and modality.
int32 neighbor_count = 3
The number of neighbors to return when querying for examples.

The Cloud Storage input instances.

Used in: Examples

ExampleGcsSource.DataFormat data_format = 1
The format in which instances are given, if not specified, assume it's JSONL format. Currently only JSONL format is supported.
optional GcsSource gcs_source = 2
The Cloud Storage location for the input instances.

The format of the input example instances.

Used in: ExampleGcsSource

DATA_FORMAT_UNSPECIFIED = 0
Format unspecified, used when unset.
JSONL = 1
Examples are stored in JSONL files.

Overrides for example-based explanations.

Used in: ExplanationSpecOverride

int32 neighbor_count = 1
The number of neighbors to return.
int32 crowding_count = 2
The number of neighbors to return that have the same crowding tag.
repeated ExamplesRestrictionsNamespace restrictions = 3
Restrict the resulting nearest neighbors to respect these constraints.
bool return_embeddings = 4
If true, return the embeddings instead of neighbors.
ExamplesOverride.DataFormat data_format = 5
The format of the data being provided with each call.

Data format enum.

Used in: ExamplesOverride

DATA_FORMAT_UNSPECIFIED = 0
Unspecified format. Must not be used.
INSTANCES = 1
Provided data is a set of model inputs.
EMBEDDINGS = 2
Provided data is a set of embeddings.

Restrictions namespace for example-based explanations overrides.

Used in: ExamplesOverride

string namespace_name = 1
The namespace name.
repeated string allow = 2
The list of allowed tags.
repeated string deny = 3
The list of deny tags.

Code generated by the model that is meant to be executed, and the result returned to the model. Generated when using the [FunctionDeclaration] tool and [FunctionCallingConfig] mode is set to [Mode.CODE].

Used in: Part

ExecutableCode.Language language = 1
Required. Programming language of the `code`.
string code = 2
Required. The code to be executed.

Supported programming languages for the generated code.

Used in: ExecutableCode

LANGUAGE_UNSPECIFIED = 0
Unspecified language. This value should not be used.
PYTHON = 1
Python >= 3.10, with numpy and simpy available.

Instance of a general execution.

Used as response type in: MetadataService.CreateExecution, MetadataService.GetExecution, MetadataService.UpdateExecution

Used as field type in: CreateExecutionRequest, LineageSubgraph, ListExecutionsResponse, PipelineTaskDetail, UpdateExecutionRequest

string name = 1
Output only. The resource name of the Execution.
string display_name = 2
User provided display name of the Execution. May be up to 128 Unicode characters.
Execution.State state = 6
The state of this Execution. This is a property of the Execution, and does not imply or capture any ongoing process. This property is managed by clients (such as Vertex AI Pipelines) and the system does not prescribe or check the validity of state transitions.
string etag = 9
An eTag used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 10
The labels with user-defined metadata to organize your Executions. 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. No more than 64 user labels can be associated with one Execution (System labels are excluded).
optional protobuf.Timestamp create_time = 11
Output only. Timestamp when this Execution was created.
optional protobuf.Timestamp update_time = 12
Output only. Timestamp when this Execution was last updated.
string schema_title = 13
The title of the schema describing the metadata. Schema title and version is expected to be registered in earlier Create Schema calls. And both are used together as unique identifiers to identify schemas within the local metadata store.
string schema_version = 14
The version of the schema in `schema_title` to use. Schema title and version is expected to be registered in earlier Create Schema calls. And both are used together as unique identifiers to identify schemas within the local metadata store.
optional protobuf.Struct metadata = 15
Properties of the Execution. Top level metadata keys' heading and trailing spaces will be trimmed. The size of this field should not exceed 200KB.
string description = 16
Description of the Execution

Describes the state of the Execution.

Used in: Execution

STATE_UNSPECIFIED = 0
Unspecified Execution state
NEW = 1
The Execution is new
RUNNING = 2
The Execution is running
COMPLETE = 3
The Execution has finished running
FAILED = 4
The Execution has failed
CACHED = 5
The Execution completed through Cache hit.
CANCELLED = 6
The Execution was cancelled.

Explanation of a prediction (provided in [PredictResponse.predictions][google.cloud.aiplatform.v1.PredictResponse.predictions]) produced by the Model on a given [instance][google.cloud.aiplatform.v1.ExplainRequest.instances].

Used in: EvaluatedAnnotationExplanation, ExplainResponse

repeated Attribution attributions = 1
Output only. Feature attributions grouped by predicted outputs. For Models that predict only one output, such as regression Models that predict only one score, there is only one attibution that explains the predicted output. For Models that predict multiple outputs, such as multiclass Models that predict multiple classes, each element explains one specific item. [Attribution.output_index][google.cloud.aiplatform.v1.Attribution.output_index] can be used to identify which output this attribution is explaining. By default, we provide Shapley values for the predicted class. However, you can configure the explanation request to generate Shapley values for any other classes too. For example, if a model predicts a probability of `0.4` for approving a loan application, the model's decision is to reject the application since `p(reject) = 0.6 > p(approve) = 0.4`, and the default Shapley values would be computed for rejection decision and not approval, even though the latter might be the positive class. If users set [ExplanationParameters.top_k][google.cloud.aiplatform.v1.ExplanationParameters.top_k], the attributions are sorted by [instance_output_value][google.cloud.aiplatform.v1.Attribution.instance_output_value] in descending order. If [ExplanationParameters.output_indices][google.cloud.aiplatform.v1.ExplanationParameters.output_indices] is specified, the attributions are stored by [Attribution.output_index][google.cloud.aiplatform.v1.Attribution.output_index] in the same order as they appear in the output_indices.
repeated Neighbor neighbors = 2
Output only. List of the nearest neighbors for example-based explanations. For models deployed with the examples explanations feature enabled, the attributions field is empty and instead the neighbors field is populated.

Metadata describing the Model's input and output for explanation.

Used in: ExplanationSpec

map<string, ExplanationMetadata.InputMetadata> inputs = 1
Required. Map from feature names to feature input metadata. Keys are the name of the features. Values are the specification of the feature. An empty InputMetadata is valid. It describes a text feature which has the name specified as the key in [ExplanationMetadata.inputs][google.cloud.aiplatform.v1.ExplanationMetadata.inputs]. The baseline of the empty feature is chosen by Vertex AI. For Vertex AI-provided Tensorflow images, the key can be any friendly name of the feature. Once specified, [featureAttributions][google.cloud.aiplatform.v1.Attribution.feature_attributions] are keyed by this key (if not grouped with another feature). For custom images, the key must match with the key in [instance][google.cloud.aiplatform.v1.ExplainRequest.instances].
map<string, ExplanationMetadata.OutputMetadata> outputs = 2
Required. Map from output names to output metadata. For Vertex AI-provided Tensorflow images, keys can be any user defined string that consists of any UTF-8 characters. For custom images, keys are the name of the output field in the prediction to be explained. Currently only one key is allowed.
string feature_attributions_schema_uri = 3
Points to a YAML file stored on Google Cloud Storage describing the format of the [feature attributions][google.cloud.aiplatform.v1.Attribution.feature_attributions]. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). AutoML tabular Models always have this field populated by Vertex AI. Note: The URI given on output may be different, including the URI scheme, than the one given on input. The output URI will point to a location where the user only has a read access.
string latent_space_source = 5
Name of the source to generate embeddings for example based explanations.

Metadata of the input of a feature. Fields other than [InputMetadata.input_baselines][google.cloud.aiplatform.v1.ExplanationMetadata.InputMetadata.input_baselines] are applicable only for Models that are using Vertex AI-provided images for Tensorflow.

Used in: ExplanationMetadata

repeated protobuf.Value input_baselines = 1
Baseline inputs for this feature. If no baseline is specified, Vertex AI chooses the baseline for this feature. If multiple baselines are specified, Vertex AI returns the average attributions across them in [Attribution.feature_attributions][google.cloud.aiplatform.v1.Attribution.feature_attributions]. For Vertex AI-provided Tensorflow images (both 1.x and 2.x), the shape of each baseline must match the shape of the input tensor. If a scalar is provided, we broadcast to the same shape as the input tensor. For custom images, the element of the baselines must be in the same format as the feature's input in the [instance][google.cloud.aiplatform.v1.ExplainRequest.instances][]. The schema of any single instance may be specified via Endpoint's DeployedModels' [Model's][google.cloud.aiplatform.v1.DeployedModel.model] [PredictSchemata's][google.cloud.aiplatform.v1.Model.predict_schemata] [instance_schema_uri][google.cloud.aiplatform.v1.PredictSchemata.instance_schema_uri].
string input_tensor_name = 2
Name of the input tensor for this feature. Required and is only applicable to Vertex AI-provided images for Tensorflow.
InputMetadata.Encoding encoding = 3
Defines how the feature is encoded into the input tensor. Defaults to IDENTITY.
string modality = 4
Modality of the feature. Valid values are: numeric, image. Defaults to numeric.
optional InputMetadata.FeatureValueDomain feature_value_domain = 5
The domain details of the input feature value. Like min/max, original mean or standard deviation if normalized.
string indices_tensor_name = 6
Specifies the index of the values of the input tensor. Required when the input tensor is a sparse representation. Refer to Tensorflow documentation for more details: https://www.tensorflow.org/api_docs/python/tf/sparse/SparseTensor.
string dense_shape_tensor_name = 7
Specifies the shape of the values of the input if the input is a sparse representation. Refer to Tensorflow documentation for more details: https://www.tensorflow.org/api_docs/python/tf/sparse/SparseTensor.
repeated string index_feature_mapping = 8
A list of feature names for each index in the input tensor. Required when the input [InputMetadata.encoding][google.cloud.aiplatform.v1.ExplanationMetadata.InputMetadata.encoding] is BAG_OF_FEATURES, BAG_OF_FEATURES_SPARSE, INDICATOR.
string encoded_tensor_name = 9
Encoded tensor is a transformation of the input tensor. Must be provided if choosing [Integrated Gradients attribution][google.cloud.aiplatform.v1.ExplanationParameters.integrated_gradients_attribution] or [XRAI attribution][google.cloud.aiplatform.v1.ExplanationParameters.xrai_attribution] and the input tensor is not differentiable. An encoded tensor is generated if the input tensor is encoded by a lookup table.
repeated protobuf.Value encoded_baselines = 10
A list of baselines for the encoded tensor. The shape of each baseline should match the shape of the encoded tensor. If a scalar is provided, Vertex AI broadcasts to the same shape as the encoded tensor.
optional InputMetadata.Visualization visualization = 11
Visualization configurations for image explanation.
string group_name = 12
Name of the group that the input belongs to. Features with the same group name will be treated as one feature when computing attributions. Features grouped together can have different shapes in value. If provided, there will be one single attribution generated in [Attribution.feature_attributions][google.cloud.aiplatform.v1.Attribution.feature_attributions], keyed by the group name.

Defines how a feature is encoded. Defaults to IDENTITY.

Used in: InputMetadata

ENCODING_UNSPECIFIED = 0
Default value. This is the same as IDENTITY.
IDENTITY = 1
The tensor represents one feature.
BAG_OF_FEATURES = 2
The tensor represents a bag of features where each index maps to a feature. [InputMetadata.index_feature_mapping][google.cloud.aiplatform.v1.ExplanationMetadata.InputMetadata.index_feature_mapping] must be provided for this encoding. For example: ``` input = [27, 6.0, 150] index_feature_mapping = ["age", "height", "weight"] ```
BAG_OF_FEATURES_SPARSE = 3
The tensor represents a bag of features where each index maps to a feature. Zero values in the tensor indicates feature being non-existent. [InputMetadata.index_feature_mapping][google.cloud.aiplatform.v1.ExplanationMetadata.InputMetadata.index_feature_mapping] must be provided for this encoding. For example: ``` input = [2, 0, 5, 0, 1] index_feature_mapping = ["a", "b", "c", "d", "e"] ```
INDICATOR = 4
The tensor is a list of binaries representing whether a feature exists or not (1 indicates existence). [InputMetadata.index_feature_mapping][google.cloud.aiplatform.v1.ExplanationMetadata.InputMetadata.index_feature_mapping] must be provided for this encoding. For example: ``` input = [1, 0, 1, 0, 1] index_feature_mapping = ["a", "b", "c", "d", "e"] ```
COMBINED_EMBEDDING = 5
The tensor is encoded into a 1-dimensional array represented by an encoded tensor. [InputMetadata.encoded_tensor_name][google.cloud.aiplatform.v1.ExplanationMetadata.InputMetadata.encoded_tensor_name] must be provided for this encoding. For example: ``` input = ["This", "is", "a", "test", "."] encoded = [0.1, 0.2, 0.3, 0.4, 0.5] ```
CONCAT_EMBEDDING = 6
Select this encoding when the input tensor is encoded into a 2-dimensional array represented by an encoded tensor. [InputMetadata.encoded_tensor_name][google.cloud.aiplatform.v1.ExplanationMetadata.InputMetadata.encoded_tensor_name] must be provided for this encoding. The first dimension of the encoded tensor's shape is the same as the input tensor's shape. For example: ``` input = ["This", "is", "a", "test", "."] encoded = [[0.1, 0.2, 0.3, 0.4, 0.5], [0.2, 0.1, 0.4, 0.3, 0.5], [0.5, 0.1, 0.3, 0.5, 0.4], [0.5, 0.3, 0.1, 0.2, 0.4], [0.4, 0.3, 0.2, 0.5, 0.1]] ```

Domain details of the input feature value. Provides numeric information about the feature, such as its range (min, max). If the feature has been pre-processed, for example with z-scoring, then it provides information about how to recover the original feature. For example, if the input feature is an image and it has been pre-processed to obtain 0-mean and stddev = 1 values, then original_mean, and original_stddev refer to the mean and stddev of the original feature (e.g. image tensor) from which input feature (with mean = 0 and stddev = 1) was obtained.

Used in: InputMetadata

float min_value = 1
The minimum permissible value for this feature.
float max_value = 2
The maximum permissible value for this feature.
float original_mean = 3
If this input feature has been normalized to a mean value of 0, the original_mean specifies the mean value of the domain prior to normalization.
float original_stddev = 4
If this input feature has been normalized to a standard deviation of 1.0, the original_stddev specifies the standard deviation of the domain prior to normalization.

Visualization configurations for image explanation.

Used in: InputMetadata

Visualization.Type type = 1
Type of the image visualization. Only applicable to [Integrated Gradients attribution][google.cloud.aiplatform.v1.ExplanationParameters.integrated_gradients_attribution]. OUTLINES shows regions of attribution, while PIXELS shows per-pixel attribution. Defaults to OUTLINES.
Visualization.Polarity polarity = 2
Whether to only highlight pixels with positive contributions, negative or both. Defaults to POSITIVE.
Visualization.ColorMap color_map = 3
The color scheme used for the highlighted areas. Defaults to PINK_GREEN for [Integrated Gradients attribution][google.cloud.aiplatform.v1.ExplanationParameters.integrated_gradients_attribution], which shows positive attributions in green and negative in pink. Defaults to VIRIDIS for [XRAI attribution][google.cloud.aiplatform.v1.ExplanationParameters.xrai_attribution], which highlights the most influential regions in yellow and the least influential in blue.
float clip_percent_upperbound = 4
Excludes attributions above the specified percentile from the highlighted areas. Using the clip_percent_upperbound and clip_percent_lowerbound together can be useful for filtering out noise and making it easier to see areas of strong attribution. Defaults to 99.9.
float clip_percent_lowerbound = 5
Excludes attributions below the specified percentile, from the highlighted areas. Defaults to 62.
Visualization.OverlayType overlay_type = 6
How the original image is displayed in the visualization. Adjusting the overlay can help increase visual clarity if the original image makes it difficult to view the visualization. Defaults to NONE.

The color scheme used for highlighting areas.

Used in: Visualization

COLOR_MAP_UNSPECIFIED = 0
Should not be used.
PINK_GREEN = 1
Positive: green. Negative: pink.
VIRIDIS = 2
Viridis color map: A perceptually uniform color mapping which is easier to see by those with colorblindness and progresses from yellow to green to blue. Positive: yellow. Negative: blue.
RED = 3
Positive: red. Negative: red.
GREEN = 4
Positive: green. Negative: green.
RED_GREEN = 6
Positive: green. Negative: red.
PINK_WHITE_GREEN = 5
PiYG palette.

How the original image is displayed in the visualization.

Used in: Visualization

OVERLAY_TYPE_UNSPECIFIED = 0
Default value. This is the same as NONE.
NONE = 1
No overlay.
ORIGINAL = 2
The attributions are shown on top of the original image.
GRAYSCALE = 3
The attributions are shown on top of grayscaled version of the original image.
MASK_BLACK = 4
The attributions are used as a mask to reveal predictive parts of the image and hide the un-predictive parts.

Whether to only highlight pixels with positive contributions, negative or both. Defaults to POSITIVE.

Used in: Visualization

POLARITY_UNSPECIFIED = 0
Default value. This is the same as POSITIVE.
POSITIVE = 1
Highlights the pixels/outlines that were most influential to the model's prediction.
NEGATIVE = 2
Setting polarity to negative highlights areas that does not lead to the models's current prediction.
BOTH = 3
Shows both positive and negative attributions.

Type of the image visualization. Only applicable to [Integrated Gradients attribution][google.cloud.aiplatform.v1.ExplanationParameters.integrated_gradients_attribution].

Used in: Visualization

TYPE_UNSPECIFIED = 0
Should not be used.
PIXELS = 1
Shows which pixel contributed to the image prediction.
OUTLINES = 2
Shows which region contributed to the image prediction by outlining the region.

Metadata of the prediction output to be explained.

Used in: ExplanationMetadata

oneof display_name_mapping
Defines how to map [Attribution.output_index][google.cloud.aiplatform.v1.Attribution.output_index] to [Attribution.output_display_name][google.cloud.aiplatform.v1.Attribution.output_display_name]. If neither of the fields are specified, [Attribution.output_display_name][google.cloud.aiplatform.v1.Attribution.output_display_name] will not be populated.
- protobuf.Value index_display_name_mapping = 1
  Static mapping between the index and display name. Use this if the outputs are a deterministic n-dimensional array, e.g. a list of scores of all the classes in a pre-defined order for a multi-classification Model. It's not feasible if the outputs are non-deterministic, e.g. the Model produces top-k classes or sort the outputs by their values. The shape of the value must be an n-dimensional array of strings. The number of dimensions must match that of the outputs to be explained. The [Attribution.output_display_name][google.cloud.aiplatform.v1.Attribution.output_display_name] is populated by locating in the mapping with [Attribution.output_index][google.cloud.aiplatform.v1.Attribution.output_index].
- string display_name_mapping_key = 2
  Specify a field name in the prediction to look for the display name. Use this if the prediction contains the display names for the outputs. The display names in the prediction must have the same shape of the outputs, so that it can be located by [Attribution.output_index][google.cloud.aiplatform.v1.Attribution.output_index] for a specific output.
string output_tensor_name = 3
Name of the output tensor. Required and is only applicable to Vertex AI provided images for Tensorflow.

The [ExplanationMetadata][google.cloud.aiplatform.v1.ExplanationMetadata] entries that can be overridden at [online explanation][google.cloud.aiplatform.v1.PredictionService.Explain] time.

Used in: ExplanationSpecOverride

map<string, ExplanationMetadataOverride.InputMetadataOverride> inputs = 1
Required. Overrides the [input metadata][google.cloud.aiplatform.v1.ExplanationMetadata.inputs] of the features. The key is the name of the feature to be overridden. The keys specified here must exist in the input metadata to be overridden. If a feature is not specified here, the corresponding feature's input metadata is not overridden.

The [input metadata][google.cloud.aiplatform.v1.ExplanationMetadata.InputMetadata] entries to be overridden.

Used in: ExplanationMetadataOverride

repeated protobuf.Value input_baselines = 1
Baseline inputs for this feature. This overrides the `input_baseline` field of the [ExplanationMetadata.InputMetadata][google.cloud.aiplatform.v1.ExplanationMetadata.InputMetadata] object of the corresponding feature's input metadata. If it's not specified, the original baselines are not overridden.

Parameters to configure explaining for Model's predictions.

Used in: ExplanationSpec, ExplanationSpecOverride

oneof method
- SampledShapleyAttribution sampled_shapley_attribution = 1
  An attribution method that approximates Shapley values for features that contribute to the label being predicted. A sampling strategy is used to approximate the value rather than considering all subsets of features. Refer to this paper for model details: https://arxiv.org/abs/1306.4265.
- IntegratedGradientsAttribution integrated_gradients_attribution = 2
  An attribution method that computes Aumann-Shapley values taking advantage of the model's fully differentiable structure. Refer to this paper for more details: https://arxiv.org/abs/1703.01365
- XraiAttribution xrai_attribution = 3
  An attribution method that redistributes Integrated Gradients attribution to segmented regions, taking advantage of the model's fully differentiable structure. Refer to this paper for more details: https://arxiv.org/abs/1906.02825 XRAI currently performs better on natural images, like a picture of a house or an animal. If the images are taken in artificial environments, like a lab or manufacturing line, or from diagnostic equipment, like x-rays or quality-control cameras, use Integrated Gradients instead.
- Examples examples = 7
  Example-based explanations that returns the nearest neighbors from the provided dataset.
int32 top_k = 4
If populated, returns attributions for top K indices of outputs (defaults to 1). Only applies to Models that predicts more than one outputs (e,g, multi-class Models). When set to -1, returns explanations for all outputs.
optional protobuf.ListValue output_indices = 5
If populated, only returns attributions that have [output_index][google.cloud.aiplatform.v1.Attribution.output_index] contained in output_indices. It must be an ndarray of integers, with the same shape of the output it's explaining. If not populated, returns attributions for [top_k][google.cloud.aiplatform.v1.ExplanationParameters.top_k] indices of outputs. If neither top_k nor output_indices is populated, returns the argmax index of the outputs. Only applicable to Models that predict multiple outputs (e,g, multi-class Models that predict multiple classes).

Specification of Model explanation.

Used in: BatchPredictionJob, DeployedModel, Model, ModelEvaluation.ModelEvaluationExplanationSpec

optional ExplanationParameters parameters = 1
Required. Parameters that configure explaining of the Model's predictions.
optional ExplanationMetadata metadata = 2
Optional. Metadata describing the Model's input and output for explanation.

The [ExplanationSpec][google.cloud.aiplatform.v1.ExplanationSpec] entries that can be overridden at [online explanation][google.cloud.aiplatform.v1.PredictionService.Explain] time.

Used in: ExplainRequest

optional ExplanationParameters parameters = 1
The parameters to be overridden. Note that the attribution method cannot be changed. If not specified, no parameter is overridden.
optional ExplanationMetadataOverride metadata = 2
The metadata to be overridden. If not specified, no metadata is overridden.
optional ExamplesOverride examples_override = 3
The example-based explanations parameter overrides.

Describes what part of the Dataset is to be exported, the destination of the export and how to export.

Used in: ExportDataRequest

oneof destination
The destination of the output.
- GcsDestination gcs_destination = 1
  The Google Cloud Storage location where the output is to be written to. In the given directory a new directory will be created with name: `export-data-<dataset-display-name>-<timestamp-of-export-call>` where timestamp is in YYYY-MM-DDThh:mm:ss.sssZ ISO-8601 format. All export output will be written into that directory. Inside that directory, annotations with the same schema will be grouped into sub directories which are named with the corresponding annotations' schema title. Inside these sub directories, a schema.yaml will be created to describe the output format.
oneof split
The instructions how the export data should be split between the training, validation and test sets.
- ExportFractionSplit fraction_split = 5
  Split based on fractions defining the size of each set.
- ExportFilterSplit filter_split = 7
  Split based on the provided filters for each set.
string annotations_filter = 2
An expression for filtering what part of the Dataset is to be exported. Only Annotations that match this filter will be exported. The filter syntax is the same as in [ListAnnotations][google.cloud.aiplatform.v1.DatasetService.ListAnnotations].
string saved_query_id = 11
The ID of a SavedQuery (annotation set) under the Dataset specified by [ExportDataRequest.name][google.cloud.aiplatform.v1.ExportDataRequest.name] used for filtering Annotations for training. Only used for custom training data export use cases. Only applicable to Datasets that have SavedQueries. Only Annotations that are associated with this SavedQuery are used in respectively training. When used in conjunction with [annotations_filter][google.cloud.aiplatform.v1.ExportDataConfig.annotations_filter], the Annotations used for training are filtered by both [saved_query_id][google.cloud.aiplatform.v1.ExportDataConfig.saved_query_id] and [annotations_filter][google.cloud.aiplatform.v1.ExportDataConfig.annotations_filter]. Only one of [saved_query_id][google.cloud.aiplatform.v1.ExportDataConfig.saved_query_id] and [annotation_schema_uri][google.cloud.aiplatform.v1.ExportDataConfig.annotation_schema_uri] should be specified as both of them represent the same thing: problem type.
string annotation_schema_uri = 12
The Cloud Storage URI that points to a YAML file describing the annotation schema. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). The schema files that can be used here are found in gs://google-cloud-aiplatform/schema/dataset/annotation/, note that the chosen schema must be consistent with [metadata][google.cloud.aiplatform.v1.Dataset.metadata_schema_uri] of the Dataset specified by [ExportDataRequest.name][google.cloud.aiplatform.v1.ExportDataRequest.name]. Only used for custom training data export use cases. Only applicable to Datasets that have DataItems and Annotations. Only Annotations that both match this schema and belong to DataItems not ignored by the split method are used in respectively training, validation or test role, depending on the role of the DataItem they are on. When used in conjunction with [annotations_filter][google.cloud.aiplatform.v1.ExportDataConfig.annotations_filter], the Annotations used for training are filtered by both [annotations_filter][google.cloud.aiplatform.v1.ExportDataConfig.annotations_filter] and [annotation_schema_uri][google.cloud.aiplatform.v1.ExportDataConfig.annotation_schema_uri].
ExportDataConfig.ExportUse export_use = 4
Indicates the usage of the exported files.

ExportUse indicates the usage of the exported files. It restricts file destination, format, annotations to be exported, whether to allow unannotated data to be exported and whether to clone files to temp Cloud Storage bucket.

Used in: ExportDataConfig

EXPORT_USE_UNSPECIFIED = 0
Regular user export.
CUSTOM_CODE_TRAINING = 6
Export for custom code training.

Runtime operation information for [DatasetService.ExportData][google.cloud.aiplatform.v1.DatasetService.ExportData].

optional GenericOperationMetadata generic_metadata = 1
The common part of the operation metadata.
string gcs_output_directory = 2
A Google Cloud Storage directory which path ends with '/'. The exported data is stored in the directory.

Response message for [DatasetService.ExportData][google.cloud.aiplatform.v1.DatasetService.ExportData].

repeated string exported_files = 1
All of the files that are exported in this export operation. For custom code training export, only three (training, validation and test) Cloud Storage paths in wildcard format are populated (for example, gs://.../training-*).
optional Model.DataStats data_stats = 2
Only present for custom code training export use case. Records data stats, i.e., train/validation/test item/annotation counts calculated during the export operation.

Details of operations that exports Features values.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for Featurestore export Feature values.

Describes exporting all historical Feature values of all entities of the EntityType between [start_time, end_time].

Used in: ExportFeatureValuesRequest

optional protobuf.Timestamp start_time = 2
Excludes Feature values with feature generation timestamp before this timestamp. If not set, retrieve oldest values kept in Feature Store. Timestamp, if present, must not have higher than millisecond precision.
optional protobuf.Timestamp end_time = 1
Exports Feature values as of this timestamp. If not set, retrieve values as of now. Timestamp, if present, must not have higher than millisecond precision.

Describes exporting the latest Feature values of all entities of the EntityType between [start_time, snapshot_time].

Used in: ExportFeatureValuesRequest

optional protobuf.Timestamp snapshot_time = 1
Exports Feature values as of this timestamp. If not set, retrieve values as of now. Timestamp, if present, must not have higher than millisecond precision.
optional protobuf.Timestamp start_time = 2
Excludes Feature values with feature generation timestamp before this timestamp. If not set, retrieve oldest values kept in Feature Store. Timestamp, if present, must not have higher than millisecond precision.

Response message for [FeaturestoreService.ExportFeatureValues][google.cloud.aiplatform.v1.FeaturestoreService.ExportFeatureValues].

(message has no fields)

Assigns input data to training, validation, and test sets based on the given filters, data pieces not matched by any filter are ignored. Currently only supported for Datasets containing DataItems. If any of the filters in this message are to match nothing, then they can be set as '-' (the minus sign). Supported only for unstructured Datasets.

Used in: ExportDataConfig

string training_filter = 1
Required. A filter on DataItems of the Dataset. DataItems that match this filter are used to train the Model. A filter with same syntax as the one used in [DatasetService.ListDataItems][google.cloud.aiplatform.v1.DatasetService.ListDataItems] may be used. If a single DataItem is matched by more than one of the FilterSplit filters, then it is assigned to the first set that applies to it in the training, validation, test order.
string validation_filter = 2
Required. A filter on DataItems of the Dataset. DataItems that match this filter are used to validate the Model. A filter with same syntax as the one used in [DatasetService.ListDataItems][google.cloud.aiplatform.v1.DatasetService.ListDataItems] may be used. If a single DataItem is matched by more than one of the FilterSplit filters, then it is assigned to the first set that applies to it in the training, validation, test order.
string test_filter = 3
Required. A filter on DataItems of the Dataset. DataItems that match this filter are used to test the Model. A filter with same syntax as the one used in [DatasetService.ListDataItems][google.cloud.aiplatform.v1.DatasetService.ListDataItems] may be used. If a single DataItem is matched by more than one of the FilterSplit filters, then it is assigned to the first set that applies to it in the training, validation, test order.

Assigns the input data to training, validation, and test sets as per the given fractions. Any of `training_fraction`, `validation_fraction` and `test_fraction` may optionally be provided, they must sum to up to 1. If the provided ones sum to less than 1, the remainder is assigned to sets as decided by Vertex AI. If none of the fractions are set, by default roughly 80% of data is used for training, 10% for validation, and 10% for test.

Used in: ExportDataConfig

double training_fraction = 1
The fraction of the input data that is to be used to train the Model.
double validation_fraction = 2
The fraction of the input data that is to be used to validate the Model.
double test_fraction = 3
The fraction of the input data that is to be used to evaluate the Model.

Details of [ModelService.ExportModel][google.cloud.aiplatform.v1.ModelService.ExportModel] operation.

optional GenericOperationMetadata generic_metadata = 1
The common part of the operation metadata.
optional ExportModelOperationMetadata.OutputInfo output_info = 2
Output only. Information further describing the output of this Model export.

Further describes the output of the ExportModel. Supplements [ExportModelRequest.OutputConfig][google.cloud.aiplatform.v1.ExportModelRequest.OutputConfig].

Used in: ExportModelOperationMetadata

string artifact_output_uri = 2
Output only. If the Model artifact is being exported to Google Cloud Storage this is the full path of the directory created, into which the Model files are being written to.
string image_output_uri = 3
Output only. If the Model image is being exported to Google Container Registry or Artifact Registry this is the full path of the image created.

Output configuration for the Model export.

Used in: ExportModelRequest

string export_format_id = 1
The ID of the format in which the Model must be exported. Each Model lists the [export formats it supports][google.cloud.aiplatform.v1.Model.supported_export_formats]. If no value is provided here, then the first from the list of the Model's supported formats is used by default.
optional GcsDestination artifact_destination = 3
The Cloud Storage location where the Model artifact is to be written to. Under the directory given as the destination a new one with name "`model-export-<model-display-name>-<timestamp-of-export-call>`", where timestamp is in YYYY-MM-DDThh:mm:ss.sssZ ISO-8601 format, will be created. Inside, the Model and any of its supporting files will be written. This field should only be set when the `exportableContent` field of the [Model.supported_export_formats] object contains `ARTIFACT`.
optional ContainerRegistryDestination image_destination = 4
The Google Container Registry or Artifact Registry uri where the Model container image will be copied to. This field should only be set when the `exportableContent` field of the [Model.supported_export_formats] object contains `IMAGE`.

Response message of [ModelService.ExportModel][google.cloud.aiplatform.v1.ModelService.ExportModel] operation.

(message has no fields)

The fact used in grounding.

Used in: AugmentPromptResponse, CorroborateContentRequest

optional string query = 1
Query that is used to retrieve this fact.
optional string title = 2
If present, it refers to the title of this fact.
optional string uri = 3
If present, this uri links to the source of the fact.
optional string summary = 4
If present, the summary/snippet of the fact.
optional double vector_distance = 5
If present, the distance between the query vector and this fact vector.
optional double score = 6
If present, according to the underlying Vector DB and the selected metric type, the score can be either the distance or the similarity between the query and the fact and its range depends on the metric type. For example, if the metric type is COSINE_DISTANCE, it represents the distance between the query and the fact. The larger the distance, the less relevant the fact is to the query. The range is [0, 2], while 0 means the most relevant and 2 means the least relevant.
optional RagChunk chunk = 7
If present, chunk properties.

Configuration for faster model deployment.

Used in: DeployedModel

bool fast_tryout_enabled = 2
If true, enable fast tryout feature for this deployed model.

Feature Metadata information. For example, color is a feature that describes an apple.

Used as response type in: FeatureRegistryService.GetFeature, FeaturestoreService.GetFeature, FeaturestoreService.UpdateFeature

Used as field type in: BatchCreateFeaturesResponse, CreateFeatureRequest, ListFeaturesResponse, SearchFeaturesResponse, UpdateFeatureRequest

string name = 1
Immutable. Name of the Feature. Format: `projects/{project}/locations/{location}/featurestores/{featurestore}/entityTypes/{entity_type}/features/{feature}` `projects/{project}/locations/{location}/featureGroups/{feature_group}/features/{feature}` The last part feature is assigned by the client. The feature can be up to 64 characters long and can consist only of ASCII Latin letters A-Z and a-z, underscore(_), and ASCII digits 0-9 starting with a letter. The value will be unique given an entity type.
string description = 2
Description of the Feature.
Feature.ValueType value_type = 3
Immutable. Only applicable for Vertex AI Feature Store (Legacy). Type of Feature value.
optional protobuf.Timestamp create_time = 4
Output only. Only applicable for Vertex AI Feature Store (Legacy). Timestamp when this EntityType was created.
optional protobuf.Timestamp update_time = 5
Output only. Only applicable for Vertex AI Feature Store (Legacy). Timestamp when this EntityType was most recently updated.
map<string, string> labels = 6
Optional. The labels with user-defined metadata to organize your Features. 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 on and examples of labels. No more than 64 user labels can be associated with one Feature (System labels are excluded)." System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
string etag = 7
Used to perform a consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
bool disable_monitoring = 12
Optional. Only applicable for Vertex AI Feature Store (Legacy). If not set, use the monitoring_config defined for the EntityType this Feature belongs to. Only Features with type ([Feature.ValueType][google.cloud.aiplatform.v1.Feature.ValueType]) BOOL, STRING, DOUBLE or INT64 can enable monitoring. If set to true, all types of data monitoring are disabled despite the config on EntityType.
repeated Feature.MonitoringStatsAnomaly monitoring_stats_anomalies = 11
Output only. Only applicable for Vertex AI Feature Store (Legacy). The list of historical stats and anomalies with specified objectives.
string version_column_name = 106
Only applicable for Vertex AI Feature Store. The name of the BigQuery Table/View column hosting data for this version. If no value is provided, will use feature_id.
string point_of_contact = 107
Entity responsible for maintaining this feature. Can be comma separated list of email addresses or URIs.

A list of historical [SnapshotAnalysis][google.cloud.aiplatform.v1.FeaturestoreMonitoringConfig.SnapshotAnalysis] or [ImportFeaturesAnalysis][google.cloud.aiplatform.v1.FeaturestoreMonitoringConfig.ImportFeaturesAnalysis] stats requested by user, sorted by [FeatureStatsAnomaly.start_time][google.cloud.aiplatform.v1.FeatureStatsAnomaly.start_time] descending.

Used in: Feature

MonitoringStatsAnomaly.Objective objective = 1
Output only. The objective for each stats.
optional FeatureStatsAnomaly feature_stats_anomaly = 2
Output only. The stats and anomalies generated at specific timestamp.

If the objective in the request is both Import Feature Analysis and Snapshot Analysis, this objective could be one of them. Otherwise, this objective should be the same as the objective in the request.

Used in: MonitoringStatsAnomaly

OBJECTIVE_UNSPECIFIED = 0
If it's OBJECTIVE_UNSPECIFIED, monitoring_stats will be empty.
IMPORT_FEATURE_ANALYSIS = 1
Stats are generated by Import Feature Analysis.
SNAPSHOT_ANALYSIS = 2
Stats are generated by Snapshot Analysis.

Only applicable for Vertex AI Legacy Feature Store. An enum representing the value type of a feature.

Used in: Feature

VALUE_TYPE_UNSPECIFIED = 0
The value type is unspecified.
BOOL = 1
Used for Feature that is a boolean.
BOOL_ARRAY = 2
Used for Feature that is a list of boolean.
DOUBLE = 3
Used for Feature that is double.
DOUBLE_ARRAY = 4
Used for Feature that is a list of double.
INT64 = 9
Used for Feature that is INT64.
INT64_ARRAY = 10
Used for Feature that is a list of INT64.
STRING = 11
Used for Feature that is string.
STRING_ARRAY = 12
Used for Feature that is a list of String.
BYTES = 13
Used for Feature that is bytes.
STRUCT = 14
Used for Feature that is struct.

Vertex AI Feature Group.

Used as response type in: FeatureRegistryService.GetFeatureGroup

Used as field type in: CreateFeatureGroupRequest, ListFeatureGroupsResponse, UpdateFeatureGroupRequest

oneof source
- FeatureGroup.BigQuery big_query = 7
  Indicates that features for this group come from BigQuery Table/View. By default treats the source as a sparse time series source. The BigQuery source table or view must have at least one entity ID column and a column named `feature_timestamp`.
string name = 1
Identifier. Name of the FeatureGroup. Format: `projects/{project}/locations/{location}/featureGroups/{featureGroup}`
optional protobuf.Timestamp create_time = 2
Output only. Timestamp when this FeatureGroup was created.
optional protobuf.Timestamp update_time = 3
Output only. Timestamp when this FeatureGroup was last updated.
string etag = 4
Optional. Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 5
Optional. The labels with user-defined metadata to organize your FeatureGroup. 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 on and examples of labels. No more than 64 user labels can be associated with one FeatureGroup(System labels are excluded)." System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
string description = 6
Optional. Description of the FeatureGroup.

Input source type for BigQuery Tables and Views.

Used in: FeatureGroup

optional BigQuerySource big_query_source = 1
Required. Immutable. The BigQuery source URI that points to either a BigQuery Table or View.
repeated string entity_id_columns = 2
Optional. Columns to construct entity_id / row keys. If not provided defaults to `entity_id`.
bool static_data_source = 3
Optional. Set if the data source is not a time-series.
optional BigQuery.TimeSeries time_series = 4
Optional. If the source is a time-series source, this can be set to control how downstream sources (ex: [FeatureView][google.cloud.aiplatform.v1.FeatureView] ) will treat time-series sources. If not set, will treat the source as a time-series source with `feature_timestamp` as timestamp column and no scan boundary.
bool dense = 5
Optional. If set, all feature values will be fetched from a single row per unique entityId including nulls. If not set, will collapse all rows for each unique entityId into a singe row with any non-null values if present, if no non-null values are present will sync null. ex: If source has schema `(entity_id, feature_timestamp, f0, f1)` and the following rows: `(e1, 2020-01-01T10:00:00.123Z, 10, 15)` `(e1, 2020-02-01T10:00:00.123Z, 20, null)` If dense is set, `(e1, 20, null)` is synced to online stores. If dense is not set, `(e1, 20, 15)` is synced to online stores.

Used in: BigQuery

string timestamp_column = 1
Optional. Column hosting timestamp values for a time-series source. Will be used to determine the latest `feature_values` for each entity. Optional. If not provided, column named `feature_timestamp` of type `TIMESTAMP` will be used.

Noise sigma by features. Noise sigma represents the standard deviation of the gaussian kernel that will be used to add noise to interpolated inputs prior to computing gradients.

Used in: SmoothGradConfig

repeated FeatureNoiseSigma.NoiseSigmaForFeature noise_sigma = 1
Noise sigma per feature. No noise is added to features that are not set.

Noise sigma for a single feature.

Used in: FeatureNoiseSigma

string name = 1
The name of the input feature for which noise sigma is provided. The features are defined in [explanation metadata inputs][google.cloud.aiplatform.v1.ExplanationMetadata.inputs].
float sigma = 2
This represents the standard deviation of the Gaussian kernel that will be used to add noise to the feature prior to computing gradients. Similar to [noise_sigma][google.cloud.aiplatform.v1.SmoothGradConfig.noise_sigma] but represents the noise added to the current feature. Defaults to 0.1.

Vertex AI Feature Online Store provides a centralized repository for serving ML features and embedding indexes at low latency. The Feature Online Store is a top-level container.

Used as response type in: FeatureOnlineStoreAdminService.GetFeatureOnlineStore

Used as field type in: CreateFeatureOnlineStoreRequest, ListFeatureOnlineStoresResponse, UpdateFeatureOnlineStoreRequest

oneof storage_type
- FeatureOnlineStore.Bigtable bigtable = 8
  Contains settings for the Cloud Bigtable instance that will be created to serve featureValues for all FeatureViews under this FeatureOnlineStore.
- FeatureOnlineStore.Optimized optimized = 12
  Contains settings for the Optimized store that will be created to serve featureValues for all FeatureViews under this FeatureOnlineStore. When choose Optimized storage type, need to set [PrivateServiceConnectConfig.enable_private_service_connect][google.cloud.aiplatform.v1.PrivateServiceConnectConfig.enable_private_service_connect] to use private endpoint. Otherwise will use public endpoint by default.
string name = 1
Identifier. Name of the FeatureOnlineStore. Format: `projects/{project}/locations/{location}/featureOnlineStores/{featureOnlineStore}`
optional protobuf.Timestamp create_time = 3
Output only. Timestamp when this FeatureOnlineStore was created.
optional protobuf.Timestamp update_time = 4
Output only. Timestamp when this FeatureOnlineStore was last updated.
string etag = 5
Optional. Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 6
Optional. The labels with user-defined metadata to organize your FeatureOnlineStore. 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 on and examples of labels. No more than 64 user labels can be associated with one FeatureOnlineStore(System labels are excluded)." System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
FeatureOnlineStore.State state = 7
Output only. State of the featureOnlineStore.
optional FeatureOnlineStore.DedicatedServingEndpoint dedicated_serving_endpoint = 10
Optional. The dedicated serving endpoint for this FeatureOnlineStore, which is different from common Vertex service endpoint.
optional EncryptionSpec encryption_spec = 13
Optional. Customer-managed encryption key spec for data storage. If set, online store will be secured by this key.
bool satisfies_pzs = 15
Output only. Reserved for future use.
bool satisfies_pzi = 16
Output only. Reserved for future use.

Used in: FeatureOnlineStore

optional Bigtable.AutoScaling auto_scaling = 1
Required. Autoscaling config applied to Bigtable Instance.

Used in: Bigtable

int32 min_node_count = 1
Required. The minimum number of nodes to scale down to. Must be greater than or equal to 1.
int32 max_node_count = 2
Required. The maximum number of nodes to scale up to. Must be greater than or equal to min_node_count, and less than or equal to 10 times of 'min_node_count'.
int32 cpu_utilization_target = 3
Optional. A percentage of the cluster's CPU capacity. Can be from 10% to 80%. When a cluster's CPU utilization exceeds the target that you have set, Bigtable immediately adds nodes to the cluster. When CPU utilization is substantially lower than the target, Bigtable removes nodes. If not set will default to 50%.

The dedicated serving endpoint for this FeatureOnlineStore. Only need to set when you choose Optimized storage type. Public endpoint is provisioned by default.

Used in: FeatureOnlineStore

string public_endpoint_domain_name = 2
Output only. This field will be populated with the domain name to use for this FeatureOnlineStore
optional PrivateServiceConnectConfig private_service_connect_config = 3
Optional. Private service connect config. The private service connection is available only for Optimized storage type, not for embedding management now. If [PrivateServiceConnectConfig.enable_private_service_connect][google.cloud.aiplatform.v1.PrivateServiceConnectConfig.enable_private_service_connect] set to true, customers will use private service connection to send request. Otherwise, the connection will set to public endpoint.
string service_attachment = 4
Output only. The name of the service attachment resource. Populated if private service connect is enabled and after FeatureViewSync is created.

Optimized storage type

Used in: FeatureOnlineStore

(message has no fields)

Possible states a featureOnlineStore can have.

Used in: FeatureOnlineStore

STATE_UNSPECIFIED = 0
Default value. This value is unused.
STABLE = 1
State when the featureOnlineStore configuration is not being updated and the fields reflect the current configuration of the featureOnlineStore. The featureOnlineStore is usable in this state.
UPDATING = 2
The state of the featureOnlineStore configuration when it is being updated. During an update, the fields reflect either the original configuration or the updated configuration of the featureOnlineStore. The featureOnlineStore is still usable in this state.

Selector for Features of an EntityType.

optional IdMatcher id_matcher = 1
Required. Matches Features based on ID.

Stats and Anomaly generated at specific timestamp for specific Feature. The start_time and end_time are used to define the time range of the dataset that current stats belongs to, e.g. prediction traffic is bucketed into prediction datasets by time window. If the Dataset is not defined by time window, start_time = end_time. Timestamp of the stats and anomalies always refers to end_time. Raw stats and anomalies are stored in stats_uri or anomaly_uri in the tensorflow defined protos. Field data_stats contains almost identical information with the raw stats in Vertex AI defined proto, for UI to display.

Used in: Feature.MonitoringStatsAnomaly, ModelMonitoringStatsAnomalies.FeatureHistoricStatsAnomalies

double score = 1
Feature importance score, only populated when cross-feature monitoring is enabled. For now only used to represent feature attribution score within range [0, 1] for [ModelDeploymentMonitoringObjectiveType.FEATURE_ATTRIBUTION_SKEW][google.cloud.aiplatform.v1.ModelDeploymentMonitoringObjectiveType.FEATURE_ATTRIBUTION_SKEW] and [ModelDeploymentMonitoringObjectiveType.FEATURE_ATTRIBUTION_DRIFT][google.cloud.aiplatform.v1.ModelDeploymentMonitoringObjectiveType.FEATURE_ATTRIBUTION_DRIFT].
string stats_uri = 3
Path of the stats file for current feature values in Cloud Storage bucket. Format: gs://<bucket_name>/<object_name>/stats. Example: gs://monitoring_bucket/feature_name/stats. Stats are stored as binary format with Protobuf message [tensorflow.metadata.v0.FeatureNameStatistics](https://github.com/tensorflow/metadata/blob/master/tensorflow_metadata/proto/v0/statistics.proto).
string anomaly_uri = 4
Path of the anomaly file for current feature values in Cloud Storage bucket. Format: gs://<bucket_name>/<object_name>/anomalies. Example: gs://monitoring_bucket/feature_name/anomalies. Stats are stored as binary format with Protobuf message Anoamlies are stored as binary format with Protobuf message [tensorflow.metadata.v0.AnomalyInfo] (https://github.com/tensorflow/metadata/blob/master/tensorflow_metadata/proto/v0/anomalies.proto).
double distribution_deviation = 5
Deviation from the current stats to baseline stats. 1. For categorical feature, the distribution distance is calculated by L-inifinity norm. 2. For numerical feature, the distribution distance is calculated by Jensen–Shannon divergence.
double anomaly_detection_threshold = 9
This is the threshold used when detecting anomalies. The threshold can be changed by user, so this one might be different from [ThresholdConfig.value][google.cloud.aiplatform.v1.ThresholdConfig.value].
optional protobuf.Timestamp start_time = 7
The start timestamp of window where stats were generated. For objectives where time window doesn't make sense (e.g. Featurestore Snapshot Monitoring), start_time is only used to indicate the monitoring intervals, so it always equals to (end_time - monitoring_interval).
optional protobuf.Timestamp end_time = 8
The end timestamp of window where stats were generated. For objectives where time window doesn't make sense (e.g. Featurestore Snapshot Monitoring), end_time indicates the timestamp of the data used to generate stats (e.g. timestamp we take snapshots for feature values).

Value for a feature.

Used in: FeatureValueList, FetchFeatureValuesResponse.FeatureNameValuePairList.FeatureNameValuePair, ReadFeatureValuesResponse.EntityView.Data, StructFieldValue, WriteFeatureValuesPayload

oneof value
Value for the feature.
- bool bool_value = 1
  Bool type feature value.
- double double_value = 2
  Double type feature value.
- int64 int64_value = 5
  Int64 feature value.
- string string_value = 6
  String feature value.
- BoolArray bool_array_value = 7
  A list of bool type feature value.
- DoubleArray double_array_value = 8
  A list of double type feature value.
- Int64Array int64_array_value = 11
  A list of int64 type feature value.
- StringArray string_array_value = 12
  A list of string type feature value.
- bytes bytes_value = 13
  Bytes feature value.
- StructValue struct_value = 15
  A struct type feature value.
optional FeatureValue.Metadata metadata = 14
Metadata of feature value.

Metadata of feature value.

Used in: FeatureValue

optional protobuf.Timestamp generate_time = 1
Feature generation timestamp. Typically, it is provided by user at feature ingestion time. If not, feature store will use the system timestamp when the data is ingested into feature store. For streaming ingestion, the time, aligned by days, must be no older than five years (1825 days) and no later than one year (366 days) in the future.

A destination location for Feature values and format.

Used in: BatchReadFeatureValuesRequest, ExportFeatureValuesRequest

oneof destination
- BigQueryDestination bigquery_destination = 1
  Output in BigQuery format. [BigQueryDestination.output_uri][google.cloud.aiplatform.v1.BigQueryDestination.output_uri] in [FeatureValueDestination.bigquery_destination][google.cloud.aiplatform.v1.FeatureValueDestination.bigquery_destination] must refer to a table.
- TFRecordDestination tfrecord_destination = 2
  Output in TFRecord format. Below are the mapping from Feature value type in Featurestore to Feature value type in TFRecord: Value type in Featurestore | Value type in TFRecord DOUBLE, DOUBLE_ARRAY | FLOAT_LIST INT64, INT64_ARRAY | INT64_LIST STRING, STRING_ARRAY, BYTES | BYTES_LIST true -> byte_string("true"), false -> byte_string("false") BOOL, BOOL_ARRAY (true, false) | BYTES_LIST
- CsvDestination csv_destination = 3
  Output in CSV format. Array Feature value types are not allowed in CSV format.

Container for list of values.

Used in: ReadFeatureValuesResponse.EntityView.Data

repeated FeatureValue values = 1
A list of feature values. All of them should be the same data type.

FeatureView is representation of values that the FeatureOnlineStore will serve based on its syncConfig.

Used as response type in: FeatureOnlineStoreAdminService.GetFeatureView

Used as field type in: CreateFeatureViewRequest, ListFeatureViewsResponse, UpdateFeatureViewRequest

oneof source
- FeatureView.BigQuerySource big_query_source = 6
  Optional. Configures how data is supposed to be extracted from a BigQuery source to be loaded onto the FeatureOnlineStore.
- FeatureView.FeatureRegistrySource feature_registry_source = 9
  Optional. Configures the features from a Feature Registry source that need to be loaded onto the FeatureOnlineStore.
- FeatureView.VertexRagSource vertex_rag_source = 18
  Optional. The Vertex RAG Source that the FeatureView is linked to.
string name = 1
Identifier. Name of the FeatureView. Format: `projects/{project}/locations/{location}/featureOnlineStores/{feature_online_store}/featureViews/{feature_view}`
optional protobuf.Timestamp create_time = 2
Output only. Timestamp when this FeatureView was created.
optional protobuf.Timestamp update_time = 3
Output only. Timestamp when this FeatureView was last updated.
string etag = 4
Optional. Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 5
Optional. The labels with user-defined metadata to organize your FeatureViews. 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 on and examples of labels. No more than 64 user labels can be associated with one FeatureOnlineStore(System labels are excluded)." System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
optional FeatureView.SyncConfig sync_config = 7
Configures when data is to be synced/updated for this FeatureView. At the end of the sync the latest featureValues for each entityId of this FeatureView are made ready for online serving.
optional FeatureView.IndexConfig index_config = 15
Optional. Configuration for index preparation for vector search. It contains the required configurations to create an index from source data, so that approximate nearest neighbor (a.k.a ANN) algorithms search can be performed during online serving.
optional FeatureView.OptimizedConfig optimized_config = 16
Optional. Configuration for FeatureView created under Optimized FeatureOnlineStore.
FeatureView.ServiceAgentType service_agent_type = 14
Optional. Service agent type used during data sync. By default, the Vertex AI Service Agent is used. When using an IAM Policy to isolate this FeatureView within a project, a separate service account should be provisioned by setting this field to `SERVICE_AGENT_TYPE_FEATURE_VIEW`. This will generate a separate service account to access the BigQuery source table.
string service_account_email = 13
Output only. A Service Account unique to this FeatureView. The role bigquery.dataViewer should be granted to this service account to allow Vertex AI Feature Store to sync data to the online store.
bool satisfies_pzs = 19
Output only. Reserved for future use.
bool satisfies_pzi = 20
Output only. Reserved for future use.

Used in: FeatureView

string uri = 1
Required. The BigQuery view URI that will be materialized on each sync trigger based on FeatureView.SyncConfig.
repeated string entity_id_columns = 2
Required. Columns to construct entity_id / row keys.

A Feature Registry source for features that need to be synced to Online Store.

Used in: FeatureView

repeated FeatureRegistrySource.FeatureGroup feature_groups = 1
Required. List of features that need to be synced to Online Store.
optional int64 project_number = 2
Optional. The project number of the parent project of the Feature Groups.

Features belonging to a single feature group that will be synced to Online Store.

Used in: FeatureRegistrySource

string feature_group_id = 1
Required. Identifier of the feature group.
repeated string feature_ids = 2
Required. Identifiers of features under the feature group.

Configuration for vector indexing.

Used in: FeatureView

oneof algorithm_config
The configuration with regard to the algorithms used for efficient search.
- IndexConfig.TreeAHConfig tree_ah_config = 6
  Optional. Configuration options for the tree-AH algorithm (Shallow tree + Asymmetric Hashing). Please refer to this paper for more details: https://arxiv.org/abs/1908.10396
- IndexConfig.BruteForceConfig brute_force_config = 7
  Optional. Configuration options for using brute force search, which simply implements the standard linear search in the database for each query. It is primarily meant for benchmarking and to generate the ground truth for approximate search.
string embedding_column = 1
Optional. Column of embedding. This column contains the source data to create index for vector search. embedding_column must be set when using vector search.
repeated string filter_columns = 2
Optional. Columns of features that're used to filter vector search results.
string crowding_column = 3
Optional. Column of crowding. This column contains crowding attribute which is a constraint on a neighbor list produced by [FeatureOnlineStoreService.SearchNearestEntities][google.cloud.aiplatform.v1.FeatureOnlineStoreService.SearchNearestEntities] to diversify search results. If [NearestNeighborQuery.per_crowding_attribute_neighbor_count][google.cloud.aiplatform.v1.NearestNeighborQuery.per_crowding_attribute_neighbor_count] is set to K in [SearchNearestEntitiesRequest][google.cloud.aiplatform.v1.SearchNearestEntitiesRequest], it's guaranteed that no more than K entities of the same crowding attribute are returned in the response.
optional int32 embedding_dimension = 4
Optional. The number of dimensions of the input embedding.
IndexConfig.DistanceMeasureType distance_measure_type = 5
Optional. The distance measure used in nearest neighbor search.

Configuration options for using brute force search.

Used in: IndexConfig

(message has no fields)

The distance measure used in nearest neighbor search.

Used in: IndexConfig

DISTANCE_MEASURE_TYPE_UNSPECIFIED = 0
Should not be set.
SQUARED_L2_DISTANCE = 1
Euclidean (L_2) Distance.
COSINE_DISTANCE = 2
Cosine Distance. Defined as 1 - cosine similarity. We strongly suggest using DOT_PRODUCT_DISTANCE + UNIT_L2_NORM instead of COSINE distance. Our algorithms have been more optimized for DOT_PRODUCT distance which, when combined with UNIT_L2_NORM, is mathematically equivalent to COSINE distance and results in the same ranking.
DOT_PRODUCT_DISTANCE = 3
Dot Product Distance. Defined as a negative of the dot product.

Configuration options for the tree-AH algorithm.

Used in: IndexConfig

optional int64 leaf_node_embedding_count = 1
Optional. Number of embeddings on each leaf node. The default value is 1000 if not set.

Configuration for FeatureViews created in Optimized FeatureOnlineStore.

Used in: FeatureView

optional AutomaticResources automatic_resources = 7
Optional. A description of resources that the FeatureView uses, which to large degree are decided by Vertex AI, and optionally allows only a modest additional configuration. If min_replica_count is not set, the default value is 2. If max_replica_count is not set, the default value is 6. The max allowed replica count is 1000.

Service agent type used during data sync.

Used in: FeatureView

SERVICE_AGENT_TYPE_UNSPECIFIED = 0
By default, the project-level Vertex AI Service Agent is enabled.
SERVICE_AGENT_TYPE_PROJECT = 1
Indicates the project-level Vertex AI Service Agent (https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) will be used during sync jobs.
SERVICE_AGENT_TYPE_FEATURE_VIEW = 2
Enable a FeatureView service account to be created by Vertex AI and output in the field `service_account_email`. This service account will be used to read from the source BigQuery table during sync.

Configuration for Sync. Only one option is set.

Used in: FeatureView

string cron = 1
Cron schedule (https://en.wikipedia.org/wiki/Cron) to launch scheduled runs. To explicitly set a timezone to the cron tab, apply a prefix in the cron tab: "CRON_TZ=${IANA_TIME_ZONE}" or "TZ=${IANA_TIME_ZONE}". The ${IANA_TIME_ZONE} may only be a valid string from IANA time zone database. For example, "CRON_TZ=America/New_York 1 * * * *", or "TZ=America/New_York 1 * * * *".
bool continuous = 2
Optional. If true, syncs the FeatureView in a continuous manner to Online Store.

A Vertex Rag source for features that need to be synced to Online Store.

Used in: FeatureView

string uri = 1
Required. The BigQuery view/table URI that will be materialized on each manual sync trigger. The table/view is expected to have the following columns and types at least: - `corpus_id` (STRING, NULLABLE/REQUIRED) - `file_id` (STRING, NULLABLE/REQUIRED) - `chunk_id` (STRING, NULLABLE/REQUIRED) - `chunk_data_type` (STRING, NULLABLE/REQUIRED) - `chunk_data` (STRING, NULLABLE/REQUIRED) - `embeddings` (FLOAT, REPEATED) - `file_original_uri` (STRING, NULLABLE/REQUIRED)
int64 rag_corpus_id = 2
Optional. The RAG corpus id corresponding to this FeatureView.

Format of the data in the Feature View.

Used in: FetchFeatureValuesRequest

FEATURE_VIEW_DATA_FORMAT_UNSPECIFIED = 0
Not set. Will be treated as the KeyValue format.
KEY_VALUE = 1
Return response data in key-value format.
PROTO_STRUCT = 2
Return response data in proto Struct format.

Lookup key for a feature view.

Used in: FetchFeatureValuesRequest, FetchFeatureValuesResponse

oneof key_oneof
- string key = 1
  String key to use for lookup.
- FeatureViewDataKey.CompositeKey composite_key = 2
  The actual Entity ID will be composed from this struct. This should match with the way ID is defined in the FeatureView spec.

ID that is comprised from several parts (columns).

Used in: FeatureViewDataKey

repeated string parts = 1
Parts to construct Entity ID. Should match with the same ID columns as defined in FeatureView in the same order.

FeatureViewSync is a representation of sync operation which copies data from data source to Feature View in Online Store.

Used as response type in: FeatureOnlineStoreAdminService.GetFeatureViewSync

Used as field type in: ListFeatureViewSyncsResponse

string name = 1
Identifier. Name of the FeatureViewSync. Format: `projects/{project}/locations/{location}/featureOnlineStores/{feature_online_store}/featureViews/{feature_view}/featureViewSyncs/{feature_view_sync}`
optional protobuf.Timestamp create_time = 2
Output only. Time when this FeatureViewSync is created. Creation of a FeatureViewSync means that the job is pending / waiting for sufficient resources but may not have started the actual data transfer yet.
optional type.Interval run_time = 5
Output only. Time when this FeatureViewSync is finished.
optional rpc.Status final_status = 4
Output only. Final status of the FeatureViewSync.
optional FeatureViewSync.SyncSummary sync_summary = 6
Output only. Summary of the sync job.
bool satisfies_pzs = 7
Output only. Reserved for future use.
bool satisfies_pzi = 8
Output only. Reserved for future use.

Summary from the Sync job. For continuous syncs, the summary is updated periodically. For batch syncs, it gets updated on completion of the sync.

Used in: FeatureViewSync

int64 row_synced = 1
Output only. Total number of rows synced.
int64 total_slot = 2
Output only. BigQuery slot milliseconds consumed for the sync job.
optional protobuf.Timestamp system_watermark_time = 5
Lower bound of the system time watermark for the sync job. This is only set for continuously syncing feature views.

Vertex AI Feature Store provides a centralized repository for organizing, storing, and serving ML features. The Featurestore is a top-level container for your features and their values.

Used as response type in: FeaturestoreService.GetFeaturestore

Used as field type in: CreateFeaturestoreRequest, ListFeaturestoresResponse, UpdateFeaturestoreRequest

string name = 1
Output only. Name of the Featurestore. Format: `projects/{project}/locations/{location}/featurestores/{featurestore}`
optional protobuf.Timestamp create_time = 3
Output only. Timestamp when this Featurestore was created.
optional protobuf.Timestamp update_time = 4
Output only. Timestamp when this Featurestore was last updated.
string etag = 5
Optional. Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 6
Optional. The labels with user-defined metadata to organize your Featurestore. 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 on and examples of labels. No more than 64 user labels can be associated with one Featurestore(System labels are excluded)." System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
optional Featurestore.OnlineServingConfig online_serving_config = 7
Optional. Config for online storage resources. The field should not co-exist with the field of `OnlineStoreReplicationConfig`. If both of it and OnlineStoreReplicationConfig are unset, the feature store will not have an online store and cannot be used for online serving.
Featurestore.State state = 8
Output only. State of the featurestore.
int32 online_storage_ttl_days = 13
Optional. TTL in days for feature values that will be stored in online serving storage. The Feature Store online storage periodically removes obsolete feature values older than `online_storage_ttl_days` since the feature generation time. Note that `online_storage_ttl_days` should be less than or equal to `offline_storage_ttl_days` for each EntityType under a featurestore. If not set, default to 4000 days
optional EncryptionSpec encryption_spec = 10
Optional. Customer-managed encryption key spec for data storage. If set, both of the online and offline data storage will be secured by this key.
bool satisfies_pzs = 14
Output only. Reserved for future use.
bool satisfies_pzi = 15
Output only. Reserved for future use.

OnlineServingConfig specifies the details for provisioning online serving resources.

Used in: Featurestore

int32 fixed_node_count = 2
The number of nodes for the online store. The number of nodes doesn't scale automatically, but you can manually update the number of nodes. If set to 0, the featurestore will not have an online store and cannot be used for online serving.
optional OnlineServingConfig.Scaling scaling = 4
Online serving scaling configuration. Only one of `fixed_node_count` and `scaling` can be set. Setting one will reset the other.

Online serving scaling configuration. If min_node_count and max_node_count are set to the same value, the cluster will be configured with the fixed number of node (no auto-scaling).

Used in: OnlineServingConfig

int32 min_node_count = 1
Required. The minimum number of nodes to scale down to. Must be greater than or equal to 1.
int32 max_node_count = 2
The maximum number of nodes to scale up to. Must be greater than min_node_count, and less than or equal to 10 times of 'min_node_count'.
int32 cpu_utilization_target = 3
Optional. The cpu utilization that the Autoscaler should be trying to achieve. This number is on a scale from 0 (no utilization) to 100 (total utilization), and is limited between 10 and 80. When a cluster's CPU utilization exceeds the target that you have set, Bigtable immediately adds nodes to the cluster. When CPU utilization is substantially lower than the target, Bigtable removes nodes. If not set or set to 0, default to 50.

Possible states a featurestore can have.

Used in: Featurestore

STATE_UNSPECIFIED = 0
Default value. This value is unused.
STABLE = 1
State when the featurestore configuration is not being updated and the fields reflect the current configuration of the featurestore. The featurestore is usable in this state.
UPDATING = 2
The state of the featurestore configuration when it is being updated. During an update, the fields reflect either the original configuration or the updated configuration of the featurestore. For example, `online_serving_config.fixed_node_count` can take minutes to update. While the update is in progress, the featurestore is in the UPDATING state, and the value of `fixed_node_count` can be the original value or the updated value, depending on the progress of the operation. Until the update completes, the actual number of nodes can still be the original value of `fixed_node_count`. The featurestore is still usable in this state.

Configuration of how features in Featurestore are monitored.

Used in: EntityType

optional FeaturestoreMonitoringConfig.SnapshotAnalysis snapshot_analysis = 1
The config for Snapshot Analysis Based Feature Monitoring.
optional FeaturestoreMonitoringConfig.ImportFeaturesAnalysis import_features_analysis = 2
The config for ImportFeatures Analysis Based Feature Monitoring.
optional FeaturestoreMonitoringConfig.ThresholdConfig numerical_threshold_config = 3
Threshold for numerical features of anomaly detection. This is shared by all objectives of Featurestore Monitoring for numerical features (i.e. Features with type ([Feature.ValueType][google.cloud.aiplatform.v1.Feature.ValueType]) DOUBLE or INT64).
optional FeaturestoreMonitoringConfig.ThresholdConfig categorical_threshold_config = 4
Threshold for categorical features of anomaly detection. This is shared by all types of Featurestore Monitoring for categorical features (i.e. Features with type ([Feature.ValueType][google.cloud.aiplatform.v1.Feature.ValueType]) BOOL or STRING).

Configuration of the Featurestore's ImportFeature Analysis Based Monitoring. This type of analysis generates statistics for values of each Feature imported by every [ImportFeatureValues][google.cloud.aiplatform.v1.FeaturestoreService.ImportFeatureValues] operation.

Used in: FeaturestoreMonitoringConfig

ImportFeaturesAnalysis.State state = 1
Whether to enable / disable / inherite default hebavior for import features analysis.
ImportFeaturesAnalysis.Baseline anomaly_detection_baseline = 2
The baseline used to do anomaly detection for the statistics generated by import features analysis.

Defines the baseline to do anomaly detection for feature values imported by each [ImportFeatureValues][google.cloud.aiplatform.v1.FeaturestoreService.ImportFeatureValues] operation.

Used in: ImportFeaturesAnalysis

BASELINE_UNSPECIFIED = 0
Should not be used.
LATEST_STATS = 1
Choose the later one statistics generated by either most recent snapshot analysis or previous import features analysis. If non of them exists, skip anomaly detection and only generate a statistics.
MOST_RECENT_SNAPSHOT_STATS = 2
Use the statistics generated by the most recent snapshot analysis if exists.
PREVIOUS_IMPORT_FEATURES_STATS = 3
Use the statistics generated by the previous import features analysis if exists.

The state defines whether to enable ImportFeature analysis.

Used in: ImportFeaturesAnalysis

STATE_UNSPECIFIED = 0
Should not be used.
DEFAULT = 1
The default behavior of whether to enable the monitoring. EntityType-level config: disabled. Feature-level config: inherited from the configuration of EntityType this Feature belongs to.
ENABLED = 2
Explicitly enables import features analysis. EntityType-level config: by default enables import features analysis for all Features under it. Feature-level config: enables import features analysis regardless of the EntityType-level config.
DISABLED = 3
Explicitly disables import features analysis. EntityType-level config: by default disables import features analysis for all Features under it. Feature-level config: disables import features analysis regardless of the EntityType-level config.

Configuration of the Featurestore's Snapshot Analysis Based Monitoring. This type of analysis generates statistics for each Feature based on a snapshot of the latest feature value of each entities every monitoring_interval.

Used in: FeaturestoreMonitoringConfig

bool disabled = 1
The monitoring schedule for snapshot analysis. For EntityType-level config: unset / disabled = true indicates disabled by default for Features under it; otherwise by default enable snapshot analysis monitoring with monitoring_interval for Features under it. Feature-level config: disabled = true indicates disabled regardless of the EntityType-level config; unset monitoring_interval indicates going with EntityType-level config; otherwise run snapshot analysis monitoring with monitoring_interval regardless of the EntityType-level config. Explicitly Disable the snapshot analysis based monitoring.
int32 monitoring_interval_days = 3
Configuration of the snapshot analysis based monitoring pipeline running interval. The value indicates number of days.
int32 staleness_days = 4
Customized export features time window for snapshot analysis. Unit is one day. Default value is 3 weeks. Minimum value is 1 day. Maximum value is 4000 days.

The config for Featurestore Monitoring threshold.

Used in: FeaturestoreMonitoringConfig

oneof threshold
- double value = 1
  Specify a threshold value that can trigger the alert. 1. For categorical feature, the distribution distance is calculated by L-inifinity norm. 2. For numerical feature, the distribution distance is calculated by Jensen–Shannon divergence. Each feature must have a non-zero threshold if they need to be monitored. Otherwise no alert will be triggered for that feature.

Response message for [FeatureOnlineStoreService.FetchFeatureValues][google.cloud.aiplatform.v1.FeatureOnlineStoreService.FetchFeatureValues]

Used as response type in: FeatureOnlineStoreService.FetchFeatureValues

Used as field type in: NearestNeighbors.Neighbor

oneof format
- FetchFeatureValuesResponse.FeatureNameValuePairList key_values = 3
  Feature values in KeyValue format.
- protobuf.Struct proto_struct = 2
  Feature values in proto Struct format.
optional FeatureViewDataKey data_key = 4
The data key associated with this response. Will only be populated for [FeatureOnlineStoreService.StreamingFetchFeatureValues][] RPCs.

Response structure in the format of key (feature name) and (feature) value pair.

Used in: FetchFeatureValuesResponse

repeated FeatureNameValuePairList.FeatureNameValuePair features = 1
List of feature names and values.

Feature name & value pair.

Used in: FeatureNameValuePairList

oneof data
- FeatureValue value = 2
  Feature value.
string name = 1
Feature short name.

URI based data.

Used in: Part

string mime_type = 1
Required. The IANA standard MIME type of the source data.
string file_uri = 2
Required. URI.

RagFile status.

Used in: RagFile

FileStatus.State state = 1
Output only. RagFile state.
string error_status = 2
Output only. Only when the `state` field is ERROR.

RagFile state.

Used in: FileStatus

STATE_UNSPECIFIED = 0
RagFile state is unspecified.
ACTIVE = 1
RagFile resource has been created and indexed successfully.
ERROR = 2
RagFile resource is in a problematic state. See `error_message` field for details.

Used in: InputDataConfig

string training_filter = 1
Required. A filter on DataItems of the Dataset. DataItems that match this filter are used to train the Model. A filter with same syntax as the one used in [DatasetService.ListDataItems][google.cloud.aiplatform.v1.DatasetService.ListDataItems] may be used. If a single DataItem is matched by more than one of the FilterSplit filters, then it is assigned to the first set that applies to it in the training, validation, test order.
string validation_filter = 2
Required. A filter on DataItems of the Dataset. DataItems that match this filter are used to validate the Model. A filter with same syntax as the one used in [DatasetService.ListDataItems][google.cloud.aiplatform.v1.DatasetService.ListDataItems] may be used. If a single DataItem is matched by more than one of the FilterSplit filters, then it is assigned to the first set that applies to it in the training, validation, test order.
string test_filter = 3
Required. A filter on DataItems of the Dataset. DataItems that match this filter are used to test the Model. A filter with same syntax as the one used in [DatasetService.ListDataItems][google.cloud.aiplatform.v1.DatasetService.ListDataItems] may be used. If a single DataItem is matched by more than one of the FilterSplit filters, then it is assigned to the first set that applies to it in the training, validation, test order.

A query to find a number of the nearest neighbors (most similar vectors) of a vector.

Used in: FindNeighborsRequest

oneof ranking
- Query.RRF rrf = 6
  Optional. Represents RRF algorithm that combines search results.
optional IndexDatapoint datapoint = 1
Required. The datapoint/vector whose nearest neighbors should be searched for.
int32 neighbor_count = 2
The number of nearest neighbors to be retrieved from database for each query. If not set, will use the default from the service configuration (https://cloud.google.com/vertex-ai/docs/matching-engine/configuring-indexes#nearest-neighbor-search-config).
int32 per_crowding_attribute_neighbor_count = 3
Crowding is a constraint on a neighbor list produced by nearest neighbor search requiring that no more than some value k' of the k neighbors returned have the same value of crowding_attribute. It's used for improving result diversity. This field is the maximum number of matches with the same crowding tag.
int32 approximate_neighbor_count = 4
The number of neighbors to find via approximate search before exact reordering is performed. If not set, the default value from scam config is used; if set, this value must be > 0.
double fraction_leaf_nodes_to_search_override = 5
The fraction of the number of leaves to search, set at query time allows user to tune search performance. This value increase result in both search accuracy and latency increase. The value should be between 0.0 and 1.0. If not set or set to 0.0, query uses the default value specified in NearestNeighborSearchConfig.TreeAHConfig.fraction_leaf_nodes_to_search.

Parameters for RRF algorithm that combines search results.

Used in: Query

float alpha = 1
Required. Users can provide an alpha value to give more weight to dense vs sparse results. For example, if the alpha is 0, we only return sparse and if the alpha is 1, we only return dense.

Nearest neighbors for one query.

Used in: FindNeighborsResponse

string id = 1
The ID of the query datapoint.
repeated Neighbor neighbors = 2
All its neighbors.

A neighbor of the query vector.

Used in: NearestNeighbors

optional IndexDatapoint datapoint = 1
The datapoint of the neighbor. Note that full datapoints are returned only when "return_full_datapoint" is set to true. Otherwise, only the "datapoint_id" and "crowding_tag" fields are populated.
double distance = 2
The distance between the neighbor and the dense embedding query.
double sparse_distance = 3
The distance between the neighbor and the query sparse_embedding.

Input for fluency metric.

optional FluencySpec metric_spec = 1
Required. Spec for fluency score metric.
optional FluencyInstance instance = 2
Required. Fluency instance.

Spec for fluency instance.

Used in: FluencyInput

optional string prediction = 1
Required. Output of the evaluated model.

Spec for fluency result.

optional float score = 1
Output only. Fluency score.
string explanation = 2
Output only. Explanation for fluency score.
optional float confidence = 3
Output only. Confidence for fluency score.

Spec for fluency score metric.

Used in: FluencyInput

int32 version = 1
Optional. Which version to use for evaluation.

Used in: InputDataConfig

double training_fraction = 1
The fraction of the input data that is to be used to train the Model.
double validation_fraction = 2
The fraction of the input data that is to be used to validate the Model.
double test_fraction = 3
The fraction of the input data that is to be used to evaluate the Model.

Input for fulfillment metric.

optional FulfillmentSpec metric_spec = 1
Required. Spec for fulfillment score metric.
optional FulfillmentInstance instance = 2
Required. Fulfillment instance.

Spec for fulfillment instance.

Used in: FulfillmentInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string instruction = 2
Required. Inference instruction prompt to compare prediction with.

Spec for fulfillment result.

Used in: BatchPredictionJob.OutputConfig, CsvDestination, CustomJobSpec, ExportDataConfig, ExportModelRequest.OutputConfig, ImportRagFilesConfig, InputDataConfig, ModelDeploymentMonitoringJob, ModelMonitoringObjectiveConfig.ExplanationConfig.ExplanationBaseline, RebaseTunedModelRequest, TFRecordDestination

optional float score = 1
Output only. Fulfillment score.
string explanation = 2
Output only. Explanation for fulfillment score.
optional float confidence = 3
Output only. Confidence for fulfillment score.

Spec for fulfillment metric.

Used in: FulfillmentInput

int32 version = 1
Optional. Which version to use for evaluation.

A predicted [FunctionCall] returned from the model that contains a string representing the [FunctionDeclaration.name] and a structured JSON object containing the parameters and their values.

Used in: Part

string name = 1
Required. The name of the function to call. Matches [FunctionDeclaration.name].
optional protobuf.Struct args = 2
Optional. Required. The function parameters and values in JSON object format. See [FunctionDeclaration.parameters] for parameter details.

Function calling config.

Used in: ToolConfig

FunctionCallingConfig.Mode mode = 1
Optional. Function calling mode.
repeated string allowed_function_names = 2
Optional. Function names to call. Only set when the Mode is ANY. Function names should match [FunctionDeclaration.name]. With mode set to ANY, model will predict a function call from the set of function names provided.

Function calling mode.

Used in: FunctionCallingConfig

MODE_UNSPECIFIED = 0
Unspecified function calling mode. This value should not be used.
AUTO = 1
Default model behavior, model decides to predict either function calls or natural language response.
ANY = 2
Model is constrained to always predicting function calls only. If "allowed_function_names" are set, the predicted function calls will be limited to any one of "allowed_function_names", else the predicted function calls will be any one of the provided "function_declarations".
NONE = 3
Model will not predict any function calls. Model behavior is same as when not passing any function declarations.

Structured representation of a function declaration as defined by the [OpenAPI 3.0 specification](https://spec.openapis.org/oas/v3.0.3). Included in this declaration are the function name, description, parameters and response type. This FunctionDeclaration is a representation of a block of code that can be used as a `Tool` by the model and executed by the client.

Used in: Tool

string name = 1
Required. The name of the function to call. Must start with a letter or an underscore. Must be a-z, A-Z, 0-9, or contain underscores, dots and dashes, with a maximum length of 64.
string description = 2
Optional. Description and purpose of the function. Model uses it to decide how and whether to call the function.
optional Schema parameters = 3
Optional. Describes the parameters to this function in JSON Schema Object format. Reflects the Open API 3.03 Parameter Object. string Key: the name of the parameter. Parameter names are case sensitive. Schema Value: the Schema defining the type used for the parameter. For function with no parameters, this can be left unset. Parameter names must start with a letter or an underscore and must only contain chars a-z, A-Z, 0-9, or underscores with a maximum length of 64. Example with 1 required and 1 optional parameter: type: OBJECT properties: param1: type: STRING param2: type: INTEGER required: - param1
optional Schema response = 4
Optional. Describes the output from this function in JSON Schema format. Reflects the Open API 3.03 Response Object. The Schema defines the type used for the response value of the function.

The result output from a [FunctionCall] that contains a string representing the [FunctionDeclaration.name] and a structured JSON object containing any output from the function is used as context to the model. This should contain the result of a [FunctionCall] made based on model prediction.

Used in: Part

string name = 1
Required. The name of the function to call. Matches [FunctionDeclaration.name] and [FunctionCall.name].
optional protobuf.Struct response = 2
Required. The function response in JSON object format. Use "output" key to specify function output and "error" key to specify error details (if any). If "output" and "error" keys are not specified, then whole "response" is treated as function output.

The Google Cloud Storage location where the output is to be written to.

string output_uri_prefix = 1
Required. Google Cloud Storage URI to output directory. If the uri doesn't end with '/', a '/' will be automatically appended. The directory is created if it doesn't exist.

The Google Cloud Storage location for the input content.

Used in: AvroSource, BatchPredictionJob.InputConfig, CsvSource, Examples.ExampleGcsSource, ImportDataConfig, ImportRagFilesConfig, ModelMonitoringObjectiveConfig.TrainingDataset, RagFile

repeated string uris = 1
Required. Google Cloud Storage URI(-s) to the input file(s). May contain wildcards. For more information on wildcards, see https://cloud.google.com/storage/docs/wildcards.

Request message for [PredictionService.GenerateContent].

Used as request type in: PredictionService.GenerateContent, PredictionService.StreamGenerateContent

string model = 5
Required. The fully qualified name of the publisher model or tuned model endpoint to use. Publisher model format: `projects/{project}/locations/{location}/publishers/*/models/*` Tuned model endpoint format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
repeated Content contents = 2
Required. The content of the current conversation with the model. For single-turn queries, this is a single instance. For multi-turn queries, this is a repeated field that contains conversation history + latest request.
optional Content system_instruction = 8
Optional. The user provided system instructions for the model. Note: only text should be used in parts and content in each part will be in a separate paragraph.
string cached_content = 9
Optional. The name of the cached content used as context to serve the prediction. Note: only used in explicit caching, where users can have control over caching (e.g. what content to cache) and enjoy guaranteed cost savings. Format: `projects/{project}/locations/{location}/cachedContents/{cachedContent}`
repeated Tool tools = 6
Optional. A list of `Tools` the model may use to generate the next response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model.
optional ToolConfig tool_config = 7
Optional. Tool config. This config is shared for all tools provided in the request.
map<string, string> labels = 10
Optional. The labels with user-defined metadata for the request. It is used for billing and reporting only. Label keys and values can be no longer than 63 characters (Unicode codepoints) and can only contain lowercase letters, numeric characters, underscores, and dashes. International characters are allowed. Label values are optional. Label keys must start with a letter.
repeated SafetySetting safety_settings = 3
Optional. Per request settings for blocking unsafe content. Enforced on GenerateContentResponse.candidates.
optional GenerationConfig generation_config = 4
Optional. Generation config.

Response message for [PredictionService.GenerateContent].

Used as response type in: PredictionService.GenerateContent, PredictionService.StreamGenerateContent

repeated Candidate candidates = 2
Output only. Generated candidates.
string model_version = 11
Output only. The model version used to generate the response.
optional protobuf.Timestamp create_time = 12
Output only. Timestamp when the request is made to the server.
string response_id = 13
Output only. response_id is used to identify each response. It is the encoding of the event_id.
optional GenerateContentResponse.PromptFeedback prompt_feedback = 3
Output only. Content filter results for a prompt sent in the request. Note: Sent only in the first stream chunk. Only happens when no candidates were generated due to content violations.
optional GenerateContentResponse.UsageMetadata usage_metadata = 4
Usage metadata about the response(s).

Content filter results for a prompt sent in the request.

Used in: GenerateContentResponse

PromptFeedback.BlockedReason block_reason = 1
Output only. Blocked reason.
repeated SafetyRating safety_ratings = 2
Output only. Safety ratings.
string block_reason_message = 3
Output only. A readable block reason message.

Blocked reason enumeration.

Used in: PromptFeedback

BLOCKED_REASON_UNSPECIFIED = 0
Unspecified blocked reason.
SAFETY = 1
Candidates blocked due to safety.
OTHER = 2
Candidates blocked due to other reason.
BLOCKLIST = 3
Candidates blocked due to the terms which are included from the terminology blocklist.
PROHIBITED_CONTENT = 4
Candidates blocked due to prohibited content.

Usage metadata about response(s).

Used in: GenerateContentResponse

int32 prompt_token_count = 1
Number of tokens in the request. When `cached_content` is set, this is still the total effective prompt size meaning this includes the number of tokens in the cached content.
int32 candidates_token_count = 2
Number of tokens in the response(s).
int32 total_token_count = 3
Total token count for prompt and response candidates.
int32 cached_content_token_count = 5
Output only. Number of tokens in the cached part in the input (the cached content).
repeated ModalityTokenCount prompt_tokens_details = 9
Output only. List of modalities that were processed in the request input.
repeated ModalityTokenCount cache_tokens_details = 10
Output only. List of modalities of the cached content in the request input.
repeated ModalityTokenCount candidates_tokens_details = 11
Output only. List of modalities that were returned in the response.

Generation config.

Used in: CountTokensRequest, GenerateContentRequest

optional float temperature = 1
Optional. Controls the randomness of predictions.
optional float top_p = 2
Optional. If specified, nucleus sampling will be used.
optional float top_k = 3
Optional. If specified, top-k sampling will be used.
optional int32 candidate_count = 4
Optional. Number of candidates to generate.
optional int32 max_output_tokens = 5
Optional. The maximum number of output tokens to generate per message.
repeated string stop_sequences = 6
Optional. Stop sequences.
optional bool response_logprobs = 18
Optional. If true, export the logprobs results in response.
optional int32 logprobs = 7
Optional. Logit probabilities.
optional float presence_penalty = 8
Optional. Positive penalties.
optional float frequency_penalty = 9
Optional. Frequency penalties.
optional int32 seed = 12
Optional. Seed.
string response_mime_type = 13
Optional. Output response mimetype of the generated candidate text. Supported mimetype: - `text/plain`: (default) Text output. - `application/json`: JSON response in the candidates. The model needs to be prompted to output the appropriate response type, otherwise the behavior is undefined. This is a preview feature.
optional Schema response_schema = 16
Optional. The `Schema` object allows the definition of input and output data types. These types can be objects, but also primitives and arrays. Represents a select subset of an [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema). If set, a compatible response_mime_type must also be set. Compatible mimetypes: `application/json`: Schema for JSON response.
optional GenerationConfig.RoutingConfig routing_config = 17
Optional. Routing configuration.
optional GenerationConfig.ThinkingConfig thinking_config = 25
Optional. Config for thinking features. An error will be returned if this field is set for models that don't support thinking.

The configuration for routing the request to a specific model.

Used in: GenerationConfig

oneof routing_config
Routing mode.
- RoutingConfig.AutoRoutingMode auto_mode = 1
  Automated routing.
- RoutingConfig.ManualRoutingMode manual_mode = 2
  Manual routing.

When automated routing is specified, the routing will be determined by the pretrained routing model and customer provided model routing preference.

Used in: RoutingConfig

optional AutoRoutingMode.ModelRoutingPreference model_routing_preference = 1
The model routing preference.

The model routing preference.

Used in: AutoRoutingMode

UNKNOWN = 0
Unspecified model routing preference.
PRIORITIZE_QUALITY = 1
Prefer higher quality over low cost.
BALANCED = 2
Balanced model routing preference.
PRIORITIZE_COST = 3
Prefer lower cost over higher quality.

When manual routing is set, the specified model will be used directly.

Used in: RoutingConfig

optional string model_name = 1
The model name to use. Only the public LLM models are accepted. e.g. 'gemini-1.5-pro-001'.

Config for thinking features.

Used in: GenerationConfig

optional int32 thinking_budget = 3
Optional. Indicates the thinking budget in tokens. This is only applied when enable_thinking is true.

Generic Metadata shared by all operations.

repeated rpc.Status partial_failures = 1
Output only. Partial failures encountered. E.g. single files that couldn't be read. This field should never exceed 20 entries. Status details field will contain standard Google Cloud error details.
optional protobuf.Timestamp create_time = 2
Output only. Time when the operation was created.
optional protobuf.Timestamp update_time = 3
Output only. Time when the operation was updated for the last time. If the operation has finished (successfully or not), this is the finish time.

Contains information about the source of the models generated from Generative AI Studio.

Used in: Model.BaseModelSource

string base_model_uri = 1
Required. The public base model URI.

Request message for [FeaturestoreService.GetFeature][google.cloud.aiplatform.v1.FeaturestoreService.GetFeature]. Request message for [FeatureRegistryService.GetFeature][google.cloud.aiplatform.v1.FeatureRegistryService.GetFeature].

Used as request type in: FeatureRegistryService.GetFeature, FeaturestoreService.GetFeature

string name = 1
Required. The name of the Feature resource. Format for entity_type as parent: `projects/{project}/locations/{location}/featurestores/{featurestore}/entityTypes/{entity_type}` Format for feature_group as parent: `projects/{project}/locations/{location}/featureGroups/{feature_group}`

The Google Drive location for the input content.

repeated GoogleDriveSource.ResourceId resource_ids = 1
Required. Google Drive resource IDs.

The type and ID of the Google Drive resource.

Used in: GoogleDriveSource

ResourceId.ResourceType resource_type = 1
Required. The type of the Google Drive resource.
string resource_id = 2
Required. The ID of the Google Drive resource.

The type of the Google Drive resource.

Used in: ResourceId

RESOURCE_TYPE_UNSPECIFIED = 0
Unspecified resource type.
RESOURCE_TYPE_FILE = 1
File resource type.
RESOURCE_TYPE_FOLDER = 2
Folder resource type.

Tool to retrieve public web data for grounding, powered by Google.

Used in: Tool

optional DynamicRetrievalConfig dynamic_retrieval_config = 2
Specifies the dynamic retrieval configuration for the given source.

Input for groundedness metric.

optional GroundednessSpec metric_spec = 1
Required. Spec for groundedness metric.
optional GroundednessInstance instance = 2
Required. Groundedness instance.

Spec for groundedness instance.

Used in: GroundednessInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string context = 2
Required. Background information provided in context used to compare against the prediction.

Spec for groundedness result.

optional float score = 1
Output only. Groundedness score.
string explanation = 2
Output only. Explanation for groundedness score.
optional float confidence = 3
Output only. Confidence for groundedness score.

Spec for groundedness metric.

Used in: GroundednessInput

int32 version = 1
Optional. Which version to use for evaluation.

Grounding chunk.

Used in: GroundingMetadata

oneof chunk_type
Chunk type.
- GroundingChunk.Web web = 1
  Grounding chunk from the web.
- GroundingChunk.RetrievedContext retrieved_context = 2
  Grounding chunk from context retrieved by the retrieval tools.

Chunk from context retrieved by the retrieval tools.

Used in: GroundingChunk

oneof context_details
Tool-specific details about the retrieved context.
- RagChunk rag_chunk = 4
  Additional context for the RAG retrieval result. This is only populated when using the RAG retrieval tool.
optional string uri = 1
URI reference of the attribution.
optional string title = 2
Title of the attribution.
optional string text = 3
Text of the attribution.

Chunk from the web.

Used in: GroundingChunk

optional string uri = 1
URI reference of the chunk.
optional string title = 2
Title of the chunk.

Metadata returned to client when grounding is enabled.

Used in: Candidate

repeated string web_search_queries = 1
Optional. Web search queries for the following-up web search.
optional SearchEntryPoint search_entry_point = 4
Optional. Google search entry for the following-up web searches.
repeated GroundingChunk grounding_chunks = 5
List of supporting references retrieved from specified grounding source.
repeated GroundingSupport grounding_supports = 6
Optional. List of grounding support.
optional RetrievalMetadata retrieval_metadata = 7
Optional. Output only. Retrieval metadata.

Grounding support.

Used in: GroundingMetadata

optional Segment segment = 1
Segment of the content this support belongs to.
repeated int32 grounding_chunk_indices = 2
A list of indices (into 'grounding_chunk') specifying the citations associated with the claim. For instance [1,3,4] means that grounding_chunk[1], grounding_chunk[3], grounding_chunk[4] are the retrieved content attributed to the claim.
repeated float confidence_scores = 3
Confidence score of the support references. Ranges from 0 to 1. 1 is the most confident. This list must have the same size as the grounding_chunk_indices.

Harm categories that will block the content.

Used in: SafetyRating, SafetySetting

HARM_CATEGORY_UNSPECIFIED = 0
The harm category is unspecified.
HARM_CATEGORY_HATE_SPEECH = 1
The harm category is hate speech.
HARM_CATEGORY_DANGEROUS_CONTENT = 2
The harm category is dangerous content.
HARM_CATEGORY_HARASSMENT = 3
The harm category is harassment.
HARM_CATEGORY_SEXUALLY_EXPLICIT = 4
The harm category is sexually explicit content.
HARM_CATEGORY_CIVIC_INTEGRITY = 5
Deprecated: Election filter is not longer supported. The harm category is civic integrity.

Represents a HyperparameterTuningJob. A HyperparameterTuningJob has a Study specification and multiple CustomJobs with identical CustomJob specification.

Used as response type in: JobService.CreateHyperparameterTuningJob, JobService.GetHyperparameterTuningJob

Used as field type in: CreateHyperparameterTuningJobRequest, ListHyperparameterTuningJobsResponse

string name = 1
Output only. Resource name of the HyperparameterTuningJob.
string display_name = 2
Required. The display name of the HyperparameterTuningJob. The name can be up to 128 characters long and can consist of any UTF-8 characters.
optional StudySpec study_spec = 4
Required. Study configuration of the HyperparameterTuningJob.
int32 max_trial_count = 5
Required. The desired total number of Trials.
int32 parallel_trial_count = 6
Required. The desired number of Trials to run in parallel.
int32 max_failed_trial_count = 7
The number of failed Trials that need to be seen before failing the HyperparameterTuningJob. If set to 0, Vertex AI decides how many Trials must fail before the whole job fails.
optional CustomJobSpec trial_job_spec = 8
Required. The spec of a trial job. The same spec applies to the CustomJobs created in all the trials.
repeated Trial trials = 9
Output only. Trials of the HyperparameterTuningJob.
JobState state = 10
Output only. The detailed state of the job.
optional protobuf.Timestamp create_time = 11
Output only. Time when the HyperparameterTuningJob was created.
optional protobuf.Timestamp start_time = 12
Output only. Time when the HyperparameterTuningJob for the first time entered the `JOB_STATE_RUNNING` state.
optional protobuf.Timestamp end_time = 13
Output only. Time when the HyperparameterTuningJob entered any of the following states: `JOB_STATE_SUCCEEDED`, `JOB_STATE_FAILED`, `JOB_STATE_CANCELLED`.
optional protobuf.Timestamp update_time = 14
Output only. Time when the HyperparameterTuningJob was most recently updated.
optional rpc.Status error = 15
Output only. Only populated when job's state is JOB_STATE_FAILED or JOB_STATE_CANCELLED.
map<string, string> labels = 16
The labels with user-defined metadata to organize HyperparameterTuningJobs. 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.
optional EncryptionSpec encryption_spec = 17
Customer-managed encryption key options for a HyperparameterTuningJob. If this is set, then all resources created by the HyperparameterTuningJob will be encrypted with the provided encryption key.
bool satisfies_pzs = 19
Output only. Reserved for future use.
bool satisfies_pzi = 20
Output only. Reserved for future use.

Matcher for Features of an EntityType by Feature ID.

Used in: FeatureSelector

repeated string ids = 1
Required. The following are accepted as `ids`: * A single-element list containing only `*`, which selects all Features in the target EntityType, or * A list containing only Feature IDs, which selects only Features with those IDs in the target EntityType.

Describes the location from where we import data into a Dataset, together with the labels that will be applied to the DataItems and the Annotations.

Used in: ImportDataRequest

oneof source
The source of the input.
- GcsSource gcs_source = 1
  The Google Cloud Storage location for the input content.
map<string, string> data_item_labels = 2
Labels that will be applied to newly imported DataItems. If an identical DataItem as one being imported already exists in the Dataset, then these labels will be appended to these of the already existing one, and if labels with identical key is imported before, the old label value will be overwritten. If two DataItems are identical in the same import data operation, the labels will be combined and if key collision happens in this case, one of the values will be picked randomly. Two DataItems are considered identical if their content bytes are identical (e.g. image bytes or pdf bytes). These labels will be overridden by Annotation labels specified inside index file referenced by [import_schema_uri][google.cloud.aiplatform.v1.ImportDataConfig.import_schema_uri], e.g. jsonl file.
map<string, string> annotation_labels = 3
Labels that will be applied to newly imported Annotations. If two Annotations are identical, one of them will be deduped. Two Annotations are considered identical if their [payload][google.cloud.aiplatform.v1.Annotation.payload], [payload_schema_uri][google.cloud.aiplatform.v1.Annotation.payload_schema_uri] and all of their [labels][google.cloud.aiplatform.v1.Annotation.labels] are the same. These labels will be overridden by Annotation labels specified inside index file referenced by [import_schema_uri][google.cloud.aiplatform.v1.ImportDataConfig.import_schema_uri], e.g. jsonl file.
string import_schema_uri = 4
Required. Points to a YAML file stored on Google Cloud Storage describing the import format. Validation will be done against the schema. The schema is defined as an [OpenAPI 3.0.2 Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject).

Runtime operation information for [DatasetService.ImportData][google.cloud.aiplatform.v1.DatasetService.ImportData].

optional GenericOperationMetadata generic_metadata = 1
The common part of the operation metadata.

Response message for [DatasetService.ImportData][google.cloud.aiplatform.v1.DatasetService.ImportData].

(message has no fields)

Details of operations that perform import Feature values.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for Featurestore import Feature values.
int64 imported_entity_count = 2
Number of entities that have been imported by the operation.
int64 imported_feature_value_count = 3
Number of Feature values that have been imported by the operation.
repeated string source_uris = 4
The source URI from where Feature values are imported.
int64 invalid_row_count = 6
The number of rows in input source that weren't imported due to either * Not having any featureValues. * Having a null entityId. * Having a null timestamp. * Not being parsable (applicable for CSV sources).
int64 timestamp_outside_retention_rows_count = 7
The number rows that weren't ingested due to having timestamps outside the retention boundary.
repeated int64 blocking_operation_ids = 8
List of ImportFeatureValues operations running under a single EntityType that are blocking this operation.

Defines the Feature value(s) to import.

Used in: ImportFeatureValuesRequest

string id = 1
Required. ID of the Feature to import values of. This Feature must exist in the target EntityType, or the request will fail.
string source_field = 2
Source column to get the Feature values from. If not set, uses the column with the same name as the Feature ID.

Response message for [FeaturestoreService.ImportFeatureValues][google.cloud.aiplatform.v1.FeaturestoreService.ImportFeatureValues].

int64 imported_entity_count = 1
Number of entities that have been imported by the operation.
int64 imported_feature_value_count = 2
Number of Feature values that have been imported by the operation.
int64 invalid_row_count = 6
The number of rows in input source that weren't imported due to either * Not having any featureValues. * Having a null entityId. * Having a null timestamp. * Not being parsable (applicable for CSV sources).
int64 timestamp_outside_retention_rows_count = 4
The number rows that weren't ingested due to having feature timestamps outside the retention boundary.

Config for importing RagFiles.

Used in: ImportRagFilesOperationMetadata, ImportRagFilesRequest

oneof import_source
The source of the import.
- GcsSource gcs_source = 2
  Google Cloud Storage location. Supports importing individual files as well as entire Google Cloud Storage directories. Sample formats: - `gs://bucket_name/my_directory/object_name/my_file.txt` - `gs://bucket_name/my_directory`
- GoogleDriveSource google_drive_source = 3
  Google Drive location. Supports importing individual files as well as Google Drive folders.
- SlackSource slack_source = 6
  Slack channels with their corresponding access tokens.
- JiraSource jira_source = 7
  Jira queries with their corresponding authentication.
- SharePointSources share_point_sources = 13
  SharePoint sources.
oneof partial_failure_sink
Optional. If provided, all partial failures are written to the sink. Deprecated. Prefer to use the `import_result_sink`.
- GcsDestination partial_failure_gcs_sink = 11
  The Cloud Storage path to write partial failures to. Deprecated. Prefer to use `import_result_gcs_sink`.
- BigQueryDestination partial_failure_bigquery_sink = 12
  The BigQuery destination to write partial failures to. It should be a bigquery table resource name (e.g. "bq://projectId.bqDatasetId.bqTableId"). The dataset must exist. If the table does not exist, it will be created with the expected schema. If the table exists, the schema will be validated and data will be added to this existing table. Deprecated. Prefer to use `import_result_bq_sink`.
oneof import_result_sink
Optional. If provided, all successfully imported files and all partial failures are written to the sink.
- GcsDestination import_result_gcs_sink = 14
  The Cloud Storage path to write import result to.
- BigQueryDestination import_result_bigquery_sink = 15
  The BigQuery destination to write import result to. It should be a bigquery table resource name (e.g. "bq://projectId.bqDatasetId.bqTableId"). The dataset must exist. If the table does not exist, it will be created with the expected schema. If the table exists, the schema will be validated and data will be added to this existing table.
optional RagFileTransformationConfig rag_file_transformation_config = 16
Specifies the transformation config for RagFiles.
optional RagFileParsingConfig rag_file_parsing_config = 8
Optional. Specifies the parsing config for RagFiles. RAG will use the default parser if this field is not set.
int32 max_embedding_requests_per_min = 5
Optional. The max number of queries per minute that this job is allowed to make to the embedding model specified on the corpus. This value is specific to this job and not shared across other import jobs. Consult the Quotas page on the project to set an appropriate value here. If unspecified, a default value of 1,000 QPM would be used.

Runtime operation information for [VertexRagDataService.ImportRagFiles][google.cloud.aiplatform.v1.VertexRagDataService.ImportRagFiles].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.
int64 rag_corpus_id = 2
The resource ID of RagCorpus that this operation is executed on.
optional ImportRagFilesConfig import_rag_files_config = 3
Output only. The config that was passed in the ImportRagFilesRequest.
int32 progress_percentage = 4
The progress percentage of the operation. Value is in the range [0, 100]. This percentage is calculated as follows: progress_percentage = 100 * (successes + failures + skips) / total

Response message for [VertexRagDataService.ImportRagFiles][google.cloud.aiplatform.v1.VertexRagDataService.ImportRagFiles].

oneof partial_failure_sink
The location into which the partial failures were written.
- string partial_failures_gcs_path = 4
  The Google Cloud Storage path into which the partial failures were written.
- string partial_failures_bigquery_table = 5
  The BigQuery table into which the partial failures were written.
int64 imported_rag_files_count = 1
The number of RagFiles that had been imported into the RagCorpus.
int64 failed_rag_files_count = 2
The number of RagFiles that had failed while importing into the RagCorpus.
int64 skipped_rag_files_count = 3
The number of RagFiles that was skipped while importing into the RagCorpus.

A representation of a collection of database items organized in a way that allows for approximate nearest neighbor (a.k.a ANN) algorithms search.

Used as response type in: IndexService.GetIndex

Used as field type in: CreateIndexRequest, ListIndexesResponse, UpdateIndexRequest

string name = 1
Output only. The resource name of the Index.
string display_name = 2
Required. The display name of the Index. The name can be up to 128 characters long and can consist of any UTF-8 characters.
string description = 3
The description of the Index.
string metadata_schema_uri = 4
Immutable. Points to a YAML file stored on Google Cloud Storage describing additional information about the Index, that is specific to it. Unset if the Index does not have any additional information. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). Note: The URI given on output will be immutable and probably different, including the URI scheme, than the one given on input. The output URI will point to a location where the user only has a read access.
optional protobuf.Value metadata = 6
An additional information about the Index; the schema of the metadata can be found in [metadata_schema][google.cloud.aiplatform.v1.Index.metadata_schema_uri].
repeated DeployedIndexRef deployed_indexes = 7
Output only. The pointers to DeployedIndexes created from this Index. An Index can be only deleted if all its DeployedIndexes had been undeployed first.
string etag = 8
Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 9
The labels with user-defined metadata to organize your Indexes. 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.
optional protobuf.Timestamp create_time = 10
Output only. Timestamp when this Index was created.
optional protobuf.Timestamp update_time = 11
Output only. Timestamp when this Index was most recently updated. This also includes any update to the contents of the Index. Note that Operations working on this Index may have their [Operations.metadata.generic_metadata.update_time] [google.cloud.aiplatform.v1.GenericOperationMetadata.update_time] a little after the value of this timestamp, yet that does not mean their results are not already reflected in the Index. Result of any successfully completed Operation on the Index is reflected in it.
optional IndexStats index_stats = 14
Output only. Stats of the index resource.
Index.IndexUpdateMethod index_update_method = 16
Immutable. The update method to use with this Index. If not set, BATCH_UPDATE will be used by default.
optional EncryptionSpec encryption_spec = 17
Immutable. Customer-managed encryption key spec for an Index. If set, this Index and all sub-resources of this Index will be secured by this key.
bool satisfies_pzs = 18
Output only. Reserved for future use.
bool satisfies_pzi = 19
Output only. Reserved for future use.

The update method of an Index.

Used in: Index

INDEX_UPDATE_METHOD_UNSPECIFIED = 0
Should not be used.
BATCH_UPDATE = 1
BatchUpdate: user can call UpdateIndex with files on Cloud Storage of Datapoints to update.
STREAM_UPDATE = 2
StreamUpdate: user can call UpsertDatapoints/DeleteDatapoints to update the Index and the updates will be applied in corresponding DeployedIndexes in nearly real-time.

A datapoint of Index.

Used in: FindNeighborsRequest.Query, FindNeighborsResponse.Neighbor, ReadIndexDatapointsResponse, UpsertDatapointsRequest

string datapoint_id = 1
Required. Unique identifier of the datapoint.
repeated float feature_vector = 2
Required. Feature embedding vector for dense index. An array of numbers with the length of [NearestNeighborSearchConfig.dimensions].
optional IndexDatapoint.SparseEmbedding sparse_embedding = 7
Optional. Feature embedding vector for sparse index.
repeated IndexDatapoint.Restriction restricts = 4
Optional. List of Restrict of the datapoint, used to perform "restricted searches" where boolean rule are used to filter the subset of the database eligible for matching. This uses categorical tokens. See: https://cloud.google.com/vertex-ai/docs/matching-engine/filtering
repeated IndexDatapoint.NumericRestriction numeric_restricts = 6
Optional. List of Restrict of the datapoint, used to perform "restricted searches" where boolean rule are used to filter the subset of the database eligible for matching. This uses numeric comparisons.
optional IndexDatapoint.CrowdingTag crowding_tag = 5
Optional. CrowdingTag of the datapoint, the number of neighbors to return in each crowding can be configured during query.

Crowding tag is a constraint on a neighbor list produced by nearest neighbor search requiring that no more than some value k' of the k neighbors returned have the same value of crowding_attribute.

Used in: IndexDatapoint

string crowding_attribute = 1
The attribute value used for crowding. The maximum number of neighbors to return per crowding attribute value (per_crowding_attribute_num_neighbors) is configured per-query. This field is ignored if per_crowding_attribute_num_neighbors is larger than the total number of neighbors to return for a given query.

This field allows restricts to be based on numeric comparisons rather than categorical tokens.

Used in: IndexDatapoint

oneof Value
The type of Value must be consistent for all datapoints with a given namespace name. This is verified at runtime.
- int64 value_int = 2
  Represents 64 bit integer.
- float value_float = 3
  Represents 32 bit float.
- double value_double = 4
  Represents 64 bit float.
string namespace = 1
The namespace of this restriction. e.g.: cost.
NumericRestriction.Operator op = 5
This MUST be specified for queries and must NOT be specified for datapoints.

Which comparison operator to use. Should be specified for queries only; specifying this for a datapoint is an error. Datapoints for which Operator is true relative to the query's Value field will be allowlisted.

Used in: NumericRestriction

OPERATOR_UNSPECIFIED = 0
Default value of the enum.
LESS = 1
Datapoints are eligible iff their value is < the query's.
LESS_EQUAL = 2
Datapoints are eligible iff their value is <= the query's.
EQUAL = 3
Datapoints are eligible iff their value is == the query's.
GREATER_EQUAL = 4
Datapoints are eligible iff their value is >= the query's.
GREATER = 5
Datapoints are eligible iff their value is > the query's.
NOT_EQUAL = 6
Datapoints are eligible iff their value is != the query's.

Restriction of a datapoint which describe its attributes(tokens) from each of several attribute categories(namespaces).

Used in: IndexDatapoint

string namespace = 1
The namespace of this restriction. e.g.: color.
repeated string allow_list = 2
The attributes to allow in this namespace. e.g.: 'red'
repeated string deny_list = 3
The attributes to deny in this namespace. e.g.: 'blue'

Feature embedding vector for sparse index. An array of numbers whose values are located in the specified dimensions.

Used in: IndexDatapoint

repeated float values = 1
Required. The list of embedding values of the sparse vector.
repeated int64 dimensions = 2
Required. The list of indexes for the embedding values of the sparse vector.

Indexes are deployed into it. An IndexEndpoint can have multiple DeployedIndexes.

Used as response type in: IndexEndpointService.GetIndexEndpoint, IndexEndpointService.UpdateIndexEndpoint

Used as field type in: CreateIndexEndpointRequest, ListIndexEndpointsResponse, UpdateIndexEndpointRequest

string name = 1
Output only. The resource name of the IndexEndpoint.
string display_name = 2
Required. The display name of the IndexEndpoint. The name can be up to 128 characters long and can consist of any UTF-8 characters.
string description = 3
The description of the IndexEndpoint.
repeated DeployedIndex deployed_indexes = 4
Output only. The indexes deployed in this endpoint.
string etag = 5
Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 6
The labels with user-defined metadata to organize your IndexEndpoints. 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.
optional protobuf.Timestamp create_time = 7
Output only. Timestamp when this IndexEndpoint was created.
optional protobuf.Timestamp update_time = 8
Output only. Timestamp when this IndexEndpoint was last updated. This timestamp is not updated when the endpoint's DeployedIndexes are updated, e.g. due to updates of the original Indexes they are the deployments of.
string network = 9
Optional. The full name of the Google Compute Engine [network](https://cloud.google.com/compute/docs/networks-and-firewalls#networks) to which the IndexEndpoint should be peered. Private services access must already be configured for the network. If left unspecified, the Endpoint is not peered with any network. [network][google.cloud.aiplatform.v1.IndexEndpoint.network] and [private_service_connect_config][google.cloud.aiplatform.v1.IndexEndpoint.private_service_connect_config] are mutually exclusive. [Format](https://cloud.google.com/compute/docs/reference/rest/v1/networks/insert): `projects/{project}/global/networks/{network}`. Where {project} is a project number, as in '12345', and {network} is network name.
bool enable_private_service_connect = 10
Optional. Deprecated: If true, expose the IndexEndpoint via private service connect. Only one of the fields, [network][google.cloud.aiplatform.v1.IndexEndpoint.network] or [enable_private_service_connect][google.cloud.aiplatform.v1.IndexEndpoint.enable_private_service_connect], can be set.
optional PrivateServiceConnectConfig private_service_connect_config = 12
Optional. Configuration for private service connect. [network][google.cloud.aiplatform.v1.IndexEndpoint.network] and [private_service_connect_config][google.cloud.aiplatform.v1.IndexEndpoint.private_service_connect_config] are mutually exclusive.
bool public_endpoint_enabled = 13
Optional. If true, the deployed index will be accessible through public endpoint.
string public_endpoint_domain_name = 14
Output only. If [public_endpoint_enabled][google.cloud.aiplatform.v1.IndexEndpoint.public_endpoint_enabled] is true, this field will be populated with the domain name to use for this index endpoint.
optional EncryptionSpec encryption_spec = 15
Immutable. Customer-managed encryption key spec for an IndexEndpoint. If set, this IndexEndpoint and all sub-resources of this IndexEndpoint will be secured by this key.
bool satisfies_pzs = 17
Output only. Reserved for future use.
bool satisfies_pzi = 18
Output only. Reserved for future use.

IndexPrivateEndpoints proto is used to provide paths for users to send requests via private endpoints (e.g. private service access, private service connect). To send request via private service access, use match_grpc_address. To send request via private service connect, use service_attachment.

Used in: DeployedIndex

string match_grpc_address = 1
Output only. The ip address used to send match gRPC requests.
string service_attachment = 2
Output only. The name of the service attachment resource. Populated if private service connect is enabled.
repeated PscAutomatedEndpoints psc_automated_endpoints = 3
Output only. PscAutomatedEndpoints is populated if private service connect is enabled if PscAutomatedConfig is set.

Stats of the Index.

Used in: Index

int64 vectors_count = 1
Output only. The number of dense vectors in the Index.
int64 sparse_vectors_count = 3
Output only. The number of sparse vectors in the Index.
int32 shards_count = 2
Output only. The number of shards in the Index.

Specifies Vertex AI owned input data to be used for training, and possibly evaluating, the Model.

Used in: TrainingPipeline

oneof split
The instructions how the input data should be split between the training, validation and test sets. If no split type is provided, the [fraction_split][google.cloud.aiplatform.v1.InputDataConfig.fraction_split] is used by default.
- FractionSplit fraction_split = 2
  Split based on fractions defining the size of each set.
- FilterSplit filter_split = 3
  Split based on the provided filters for each set.
- PredefinedSplit predefined_split = 4
  Supported only for tabular Datasets. Split based on a predefined key.
- TimestampSplit timestamp_split = 5
  Supported only for tabular Datasets. Split based on the timestamp of the input data pieces.
- StratifiedSplit stratified_split = 12
  Supported only for tabular Datasets. Split based on the distribution of the specified column.
oneof destination
Only applicable to Custom and Hyperparameter Tuning TrainingPipelines. The destination of the training data to be written to. Supported destination file formats: * For non-tabular data: "jsonl". * For tabular data: "csv" and "bigquery". The following Vertex AI environment variables are passed to containers or python modules of the training task when this field is set: * AIP_DATA_FORMAT : Exported data format. * AIP_TRAINING_DATA_URI : Sharded exported training data uris. * AIP_VALIDATION_DATA_URI : Sharded exported validation data uris. * AIP_TEST_DATA_URI : Sharded exported test data uris.
- GcsDestination gcs_destination = 8
  The Cloud Storage location where the training data is to be written to. In the given directory a new directory is created with name: `dataset-<dataset-id>-<annotation-type>-<timestamp-of-training-call>` where timestamp is in YYYY-MM-DDThh:mm:ss.sssZ ISO-8601 format. All training input data is written into that directory. The Vertex AI environment variables representing Cloud Storage data URIs are represented in the Cloud Storage wildcard format to support sharded data. e.g.: "gs://.../training-*.jsonl" * AIP_DATA_FORMAT = "jsonl" for non-tabular data, "csv" for tabular data * AIP_TRAINING_DATA_URI = "gcs_destination/dataset-<dataset-id>-<annotation-type>-<time>/training-*.${AIP_DATA_FORMAT}" * AIP_VALIDATION_DATA_URI = "gcs_destination/dataset-<dataset-id>-<annotation-type>-<time>/validation-*.${AIP_DATA_FORMAT}" * AIP_TEST_DATA_URI = "gcs_destination/dataset-<dataset-id>-<annotation-type>-<time>/test-*.${AIP_DATA_FORMAT}"
- BigQueryDestination bigquery_destination = 10
  Only applicable to custom training with tabular Dataset with BigQuery source. The BigQuery project location where the training data is to be written to. In the given project a new dataset is created with name `dataset_<dataset-id>_<annotation-type>_<timestamp-of-training-call>` where timestamp is in YYYY_MM_DDThh_mm_ss_sssZ format. All training input data is written into that dataset. In the dataset three tables are created, `training`, `validation` and `test`. * AIP_DATA_FORMAT = "bigquery". * AIP_TRAINING_DATA_URI = "bigquery_destination.dataset_<dataset-id>_<annotation-type>_<time>.training" * AIP_VALIDATION_DATA_URI = "bigquery_destination.dataset_<dataset-id>_<annotation-type>_<time>.validation" * AIP_TEST_DATA_URI = "bigquery_destination.dataset_<dataset-id>_<annotation-type>_<time>.test"
string dataset_id = 1
Required. The ID of the Dataset in the same Project and Location which data will be used to train the Model. The Dataset must use schema compatible with Model being trained, and what is compatible should be described in the used TrainingPipeline's [training_task_definition] [google.cloud.aiplatform.v1.TrainingPipeline.training_task_definition]. For tabular Datasets, all their data is exported to training, to pick and choose from.
string annotations_filter = 6
Applicable only to Datasets that have DataItems and Annotations. A filter on Annotations of the Dataset. Only Annotations that both match this filter and belong to DataItems not ignored by the split method are used in respectively training, validation or test role, depending on the role of the DataItem they are on (for the auto-assigned that role is decided by Vertex AI). A filter with same syntax as the one used in [ListAnnotations][google.cloud.aiplatform.v1.DatasetService.ListAnnotations] may be used, but note here it filters across all Annotations of the Dataset, and not just within a single DataItem.
string annotation_schema_uri = 9
Applicable only to custom training with Datasets that have DataItems and Annotations. Cloud Storage URI that points to a YAML file describing the annotation schema. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). The schema files that can be used here are found in gs://google-cloud-aiplatform/schema/dataset/annotation/ , note that the chosen schema must be consistent with [metadata][google.cloud.aiplatform.v1.Dataset.metadata_schema_uri] of the Dataset specified by [dataset_id][google.cloud.aiplatform.v1.InputDataConfig.dataset_id]. Only Annotations that both match this schema and belong to DataItems not ignored by the split method are used in respectively training, validation or test role, depending on the role of the DataItem they are on. When used in conjunction with [annotations_filter][google.cloud.aiplatform.v1.InputDataConfig.annotations_filter], the Annotations used for training are filtered by both [annotations_filter][google.cloud.aiplatform.v1.InputDataConfig.annotations_filter] and [annotation_schema_uri][google.cloud.aiplatform.v1.InputDataConfig.annotation_schema_uri].
string saved_query_id = 7
Only applicable to Datasets that have SavedQueries. The ID of a SavedQuery (annotation set) under the Dataset specified by [dataset_id][google.cloud.aiplatform.v1.InputDataConfig.dataset_id] used for filtering Annotations for training. Only Annotations that are associated with this SavedQuery are used in respectively training. When used in conjunction with [annotations_filter][google.cloud.aiplatform.v1.InputDataConfig.annotations_filter], the Annotations used for training are filtered by both [saved_query_id][google.cloud.aiplatform.v1.InputDataConfig.saved_query_id] and [annotations_filter][google.cloud.aiplatform.v1.InputDataConfig.annotations_filter]. Only one of [saved_query_id][google.cloud.aiplatform.v1.InputDataConfig.saved_query_id] and [annotation_schema_uri][google.cloud.aiplatform.v1.InputDataConfig.annotation_schema_uri] should be specified as both of them represent the same thing: problem type.
bool persist_ml_use_assignment = 11
Whether to persist the ML use assignment to data item system labels.

A list of int64 values.

Used in: FeatureValue

repeated int64 values = 1
A list of int64 values.

An attribution method that computes the Aumann-Shapley value taking advantage of the model's fully differentiable structure. Refer to this paper for more details: https://arxiv.org/abs/1703.01365

Used in: ExplanationParameters

int32 step_count = 1
Required. The number of steps for approximating the path integral. A good value to start is 50 and gradually increase until the sum to diff property is within the desired error range. Valid range of its value is [1, 100], inclusively.
optional SmoothGradConfig smooth_grad_config = 2
Config for SmoothGrad approximation of gradients. When enabled, the gradients are approximated by averaging the gradients from noisy samples in the vicinity of the inputs. Adding noise can help improve the computed gradients. Refer to this paper for more details: https://arxiv.org/pdf/1706.03825.pdf
optional BlurBaselineConfig blur_baseline_config = 3
Config for IG with blur baseline. When enabled, a linear path from the maximally blurred image to the input image is created. Using a blurred baseline instead of zero (black image) is motivated by the BlurIG approach explained here: https://arxiv.org/abs/2004.03383

The Jira source for the ImportRagFilesRequest.

repeated JiraSource.JiraQueries jira_queries = 1
Required. The Jira queries.

JiraQueries contains the Jira queries and corresponding authentication.

Used in: JiraSource

repeated string projects = 3
A list of Jira projects to import in their entirety.
repeated string custom_queries = 4
A list of custom Jira queries to import. For information about JQL (Jira Query Language), see https://support.atlassian.com/jira-service-management-cloud/docs/use-advanced-search-with-jira-query-language-jql/
string email = 5
Required. The Jira email address.
string server_uri = 6
Required. The Jira server URI.
optional ApiAuth.ApiKeyConfig api_key_config = 7
Required. The SecretManager secret version resource name (e.g. projects/{project}/secrets/{secret}/versions/{version}) storing the Jira API key. See [Manage API tokens for your Atlassian account](https://support.atlassian.com/atlassian-account/docs/manage-api-tokens-for-your-atlassian-account/).

Describes the state of a job.

Used in: BatchPredictionJob, CustomJob, DataLabelingJob, HyperparameterTuningJob, ModelDeploymentMonitoringJob, NasJob, NotebookExecutionJob, TuningJob

JOB_STATE_UNSPECIFIED = 0
The job state is unspecified.
JOB_STATE_QUEUED = 1
The job has been just created or resumed and processing has not yet begun.
JOB_STATE_PENDING = 2
The service is preparing to run the job.
JOB_STATE_RUNNING = 3
The job is in progress.
JOB_STATE_SUCCEEDED = 4
The job completed successfully.
JOB_STATE_FAILED = 5
The job failed.
JOB_STATE_CANCELLING = 6
The job is being cancelled. From this state the job may only go to either `JOB_STATE_SUCCEEDED`, `JOB_STATE_FAILED` or `JOB_STATE_CANCELLED`.
JOB_STATE_CANCELLED = 7
The job has been cancelled.
JOB_STATE_PAUSED = 8
The job has been stopped, and can be resumed.
JOB_STATE_EXPIRED = 9
The job has expired.
JOB_STATE_UPDATING = 10
The job is being updated. Only jobs in the `RUNNING` state can be updated. After updating, the job goes back to the `RUNNING` state.
JOB_STATE_PARTIALLY_SUCCEEDED = 11
The job is partially succeeded, some results may be missing due to errors.

Contains information about the Large Model.

Used in: PublisherModel.CallToAction.Deploy

string name = 1
Required. The unique name of the large Foundation or pre-built model. Like "chat-bison", "text-bison". Or model name with version ID, like "chat-bison@001", "text-bison@005", etc.

A subgraph of the overall lineage graph. Event edges connect Artifact and Execution nodes.

Used as response type in: MetadataService.QueryArtifactLineageSubgraph, MetadataService.QueryContextLineageSubgraph, MetadataService.QueryExecutionInputsAndOutputs

repeated Artifact artifacts = 1
The Artifact nodes in the subgraph.
repeated Execution executions = 2
The Execution nodes in the subgraph.
repeated Event events = 3
The Event edges between Artifacts and Executions in the subgraph.

Request message for [FeaturestoreService.ListFeatures][google.cloud.aiplatform.v1.FeaturestoreService.ListFeatures]. Request message for [FeatureRegistryService.ListFeatures][google.cloud.aiplatform.v1.FeatureRegistryService.ListFeatures].

Used as request type in: FeatureRegistryService.ListFeatures, FeaturestoreService.ListFeatures

string parent = 1
Required. The resource name of the Location to list Features. Format for entity_type as parent: `projects/{project}/locations/{location}/featurestores/{featurestore}/entityTypes/{entity_type}` Format for feature_group as parent: `projects/{project}/locations/{location}/featureGroups/{feature_group}`
string filter = 2
Lists the Features that match the filter expression. The following filters are supported: * `value_type`: Supports = and != comparisons. * `create_time`: Supports =, !=, <, >, >=, and <= comparisons. Values must be in RFC 3339 format. * `update_time`: Supports =, !=, <, >, >=, and <= comparisons. Values must be in RFC 3339 format. * `labels`: Supports key-value equality as well as key presence. Examples: * `value_type = DOUBLE` --> Features whose type is DOUBLE. * `create_time > \"2020-01-31T15:30:00.000000Z\" OR update_time > \"2020-01-31T15:30:00.000000Z\"` --> EntityTypes created or updated after 2020-01-31T15:30:00.000000Z. * `labels.active = yes AND labels.env = prod` --> Features having both (active: yes) and (env: prod) labels. * `labels.env: *` --> Any Feature which has a label with 'env' as the key.
int32 page_size = 3
The maximum number of Features to return. The service may return fewer than this value. If unspecified, at most 1000 Features will be returned. The maximum value is 1000; any value greater than 1000 will be coerced to 1000.
string page_token = 4
A page token, received from a previous [FeaturestoreService.ListFeatures][google.cloud.aiplatform.v1.FeaturestoreService.ListFeatures] call or [FeatureRegistryService.ListFeatures][google.cloud.aiplatform.v1.FeatureRegistryService.ListFeatures] call. Provide this to retrieve the subsequent page. When paginating, all other parameters provided to [FeaturestoreService.ListFeatures][google.cloud.aiplatform.v1.FeaturestoreService.ListFeatures] or [FeatureRegistryService.ListFeatures][google.cloud.aiplatform.v1.FeatureRegistryService.ListFeatures] must match the call that provided the page token.
string order_by = 5
A comma-separated list of fields to order by, sorted in ascending order. Use "desc" after a field name for descending. Supported fields: * `feature_id` * `value_type` (Not supported for FeatureRegistry Feature) * `create_time` * `update_time`
optional protobuf.FieldMask read_mask = 6
Mask specifying which fields to read.
int32 latest_stats_count = 7
Only applicable for Vertex AI Feature Store (Legacy). If set, return the most recent [ListFeaturesRequest.latest_stats_count][google.cloud.aiplatform.v1.ListFeaturesRequest.latest_stats_count] of stats for each Feature in response. Valid value is [0, 10]. If number of stats exists < [ListFeaturesRequest.latest_stats_count][google.cloud.aiplatform.v1.ListFeaturesRequest.latest_stats_count], return all existing stats.

Response message for [FeaturestoreService.ListFeatures][google.cloud.aiplatform.v1.FeaturestoreService.ListFeatures]. Response message for [FeatureRegistryService.ListFeatures][google.cloud.aiplatform.v1.FeatureRegistryService.ListFeatures].

Used as response type in: FeatureRegistryService.ListFeatures, FeaturestoreService.ListFeatures

repeated Feature features = 1
The Features matching the request.
string next_page_token = 2
A token, which can be sent as [ListFeaturesRequest.page_token][google.cloud.aiplatform.v1.ListFeaturesRequest.page_token] to retrieve the next page. If this field is omitted, there are no subsequent pages.

Logprobs Result

Used in: Candidate

repeated LogprobsResult.TopCandidates top_candidates = 1
Length = total number of decoding steps.
repeated LogprobsResult.Candidate chosen_candidates = 2
Length = total number of decoding steps. The chosen candidates may or may not be in top_candidates.

Candidate for the logprobs token and score.

Used in: LogprobsResult, TopCandidates

optional string token = 1
The candidate’s token string value.
optional int32 token_id = 3
The candidate’s token id value.
optional float log_probability = 2
The candidate's log probability.

Candidates with top log probabilities at each decoding step.

Used in: LogprobsResult

repeated Candidate candidates = 1
Sorted by log probability in descending order.

Specification of a single machine.

Used in: BatchDedicatedResources, DedicatedResources, NotebookExecutionJob.CustomEnvironmentSpec, NotebookRuntime, NotebookRuntimeTemplate, ResourcePool, WorkerPoolSpec

string machine_type = 1
Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For [DeployedModel][google.cloud.aiplatform.v1.DeployedModel] this field is optional, and the default value is `n1-standard-2`. For [BatchPredictionJob][google.cloud.aiplatform.v1.BatchPredictionJob] or as part of [WorkerPoolSpec][google.cloud.aiplatform.v1.WorkerPoolSpec] this field is required.
AcceleratorType accelerator_type = 2
Immutable. The type of accelerator(s) that may be attached to the machine as per [accelerator_count][google.cloud.aiplatform.v1.MachineSpec.accelerator_count].
int32 accelerator_count = 3
The number of accelerators to attach to the machine.
string tpu_topology = 4
Immutable. The topology of the TPUs. Corresponds to the TPU topologies available from GKE. (Example: tpu_topology: "2x2x1").
optional ReservationAffinity reservation_affinity = 5
Optional. Immutable. Configuration controlling how this resource pool consumes reservation.

Manual batch tuning parameters.

Used in: BatchPredictionJob

int32 batch_size = 1
Immutable. The number of the records (e.g. instances) of the operation given in each batch to a machine replica. Machine type, and size of a single record should be considered when setting this parameter, higher value speeds up the batch operation's execution, but too high value will result in a whole batch not fitting in a machine's memory, and the whole operation will fail. The default value is 64.

A message representing a Measurement of a Trial. A Measurement contains the Metrics got by executing a Trial using suggested hyperparameter values.

Used in: AddTrialMeasurementRequest, CompleteTrialRequest, NasTrial, Trial

optional protobuf.Duration elapsed_duration = 1
Output only. Time that the Trial has been running at the point of this Measurement.
int64 step_count = 2
Output only. The number of steps the machine learning model has been trained for. Must be non-negative.
repeated Measurement.Metric metrics = 3
Output only. A list of metrics got by evaluating the objective functions using suggested Parameter values.

A message representing a metric in the measurement.

Used in: Measurement

string metric_id = 1
Output only. The ID of the Metric. The Metric should be defined in [StudySpec's Metrics][google.cloud.aiplatform.v1.StudySpec.metrics].
double value = 2
Output only. The value for this metric.

Instance of a general MetadataSchema.

Used as response type in: MetadataService.CreateMetadataSchema, MetadataService.GetMetadataSchema

Used as field type in: CreateMetadataSchemaRequest, ListMetadataSchemasResponse

string name = 1
Output only. The resource name of the MetadataSchema.
string schema_version = 2
The version of the MetadataSchema. The version's format must match the following regular expression: `^[0-9]+[.][0-9]+[.][0-9]+$`, which would allow to order/compare different versions. Example: 1.0.0, 1.0.1, etc.
string schema = 3
Required. The raw YAML string representation of the MetadataSchema. The combination of [MetadataSchema.version] and the schema name given by `title` in [MetadataSchema.schema] must be unique within a MetadataStore. The schema is defined as an OpenAPI 3.0.2 [MetadataSchema Object](https://github.com/OAI/OpenAPI-Specification/blob/master/versions/3.0.2.md#schemaObject)
MetadataSchema.MetadataSchemaType schema_type = 4
The type of the MetadataSchema. This is a property that identifies which metadata types will use the MetadataSchema.
optional protobuf.Timestamp create_time = 5
Output only. Timestamp when this MetadataSchema was created.
string description = 6
Description of the Metadata Schema

Describes the type of the MetadataSchema.

Used in: MetadataSchema

METADATA_SCHEMA_TYPE_UNSPECIFIED = 0
Unspecified type for the MetadataSchema.
ARTIFACT_TYPE = 1
A type indicating that the MetadataSchema will be used by Artifacts.
EXECUTION_TYPE = 2
A typee indicating that the MetadataSchema will be used by Executions.
CONTEXT_TYPE = 3
A state indicating that the MetadataSchema will be used by Contexts.

Instance of a metadata store. Contains a set of metadata that can be queried.

Used as response type in: MetadataService.GetMetadataStore

Used as field type in: CreateMetadataStoreRequest, ListMetadataStoresResponse

string name = 1
Output only. The resource name of the MetadataStore instance.
optional protobuf.Timestamp create_time = 3
Output only. Timestamp when this MetadataStore was created.
optional protobuf.Timestamp update_time = 4
Output only. Timestamp when this MetadataStore was last updated.
optional EncryptionSpec encryption_spec = 5
Customer-managed encryption key spec for a Metadata Store. If set, this Metadata Store and all sub-resources of this Metadata Store are secured using this key.
string description = 6
Description of the MetadataStore.
optional MetadataStore.MetadataStoreState state = 7
Output only. State information of the MetadataStore.
optional MetadataStore.DataplexConfig dataplex_config = 8
Optional. Dataplex integration settings.

Represents Dataplex integration settings.

Used in: MetadataStore

bool enabled_pipelines_lineage = 1
Optional. Whether or not Data Lineage synchronization is enabled for Vertex Pipelines.

Represents state information for a MetadataStore.

Used in: MetadataStore

int64 disk_utilization_bytes = 1
The disk utilization of the MetadataStore in bytes.

Input for MetricX metric.

optional MetricxSpec metric_spec = 1
Required. Spec for Metricx metric.
optional MetricxInstance instance = 2
Required. Metricx instance.

Spec for MetricX instance - The fields used for evaluation are dependent on the MetricX version.

Used in: MetricxInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Optional. Ground truth used to compare against the prediction.
optional string source = 3
Optional. Source text in original language.

Spec for MetricX result - calculates the MetricX score for the given instance using the version specified in the spec.

optional float score = 1
Output only. MetricX score. Range depends on version.

Spec for MetricX metric.

Used in: MetricxInput

optional MetricxSpec.MetricxVersion version = 1
Required. Which version to use for evaluation.
string source_language = 2
Optional. Source language in BCP-47 format.
string target_language = 3
Optional. Target language in BCP-47 format. Covers both prediction and reference.

MetricX Version options.

Used in: MetricxSpec

METRICX_VERSION_UNSPECIFIED = 0
MetricX version unspecified.
METRICX_24_REF = 1
MetricX 2024 (2.6) for translation + reference (reference-based).
METRICX_24_SRC = 2
MetricX 2024 (2.6) for translation + source (QE).
METRICX_24_SRC_REF = 3
MetricX 2024 (2.6) for translation + source + reference (source-reference-combined).

Represents one resource that exists in automl.googleapis.com, datalabeling.googleapis.com or ml.googleapis.com.

Used in: MigrateResourceResponse, SearchMigratableResourcesResponse

oneof resource
- MigratableResource.MlEngineModelVersion ml_engine_model_version = 1
  Output only. Represents one Version in ml.googleapis.com.
- MigratableResource.AutomlModel automl_model = 2
  Output only. Represents one Model in automl.googleapis.com.
- MigratableResource.AutomlDataset automl_dataset = 3
  Output only. Represents one Dataset in automl.googleapis.com.
- MigratableResource.DataLabelingDataset data_labeling_dataset = 4
  Output only. Represents one Dataset in datalabeling.googleapis.com.
optional protobuf.Timestamp last_migrate_time = 5
Output only. Timestamp when the last migration attempt on this MigratableResource started. Will not be set if there's no migration attempt on this MigratableResource.
optional protobuf.Timestamp last_update_time = 6
Output only. Timestamp when this MigratableResource was last updated.

Represents one Dataset in automl.googleapis.com.

Used in: MigratableResource

string dataset = 1
Full resource name of automl Dataset. Format: `projects/{project}/locations/{location}/datasets/{dataset}`.
string dataset_display_name = 4
The Dataset's display name in automl.googleapis.com.

Represents one Model in automl.googleapis.com.

Used in: MigratableResource

string model = 1
Full resource name of automl Model. Format: `projects/{project}/locations/{location}/models/{model}`.
string model_display_name = 3
The Model's display name in automl.googleapis.com.

Represents one Dataset in datalabeling.googleapis.com.

Used in: MigratableResource

string dataset = 1
Full resource name of data labeling Dataset. Format: `projects/{project}/datasets/{dataset}`.
string dataset_display_name = 4
The Dataset's display name in datalabeling.googleapis.com.
repeated DataLabelingDataset.DataLabelingAnnotatedDataset data_labeling_annotated_datasets = 3
The migratable AnnotatedDataset in datalabeling.googleapis.com belongs to the data labeling Dataset.

Represents one AnnotatedDataset in datalabeling.googleapis.com.

Used in: DataLabelingDataset

string annotated_dataset = 1
Full resource name of data labeling AnnotatedDataset. Format: `projects/{project}/datasets/{dataset}/annotatedDatasets/{annotated_dataset}`.
string annotated_dataset_display_name = 3
The AnnotatedDataset's display name in datalabeling.googleapis.com.

Represents one model Version in ml.googleapis.com.

Used in: MigratableResource

string endpoint = 1
The ml.googleapis.com endpoint that this model Version currently lives in. Example values: * ml.googleapis.com * us-centrall-ml.googleapis.com * europe-west4-ml.googleapis.com * asia-east1-ml.googleapis.com
string version = 2
Full resource name of ml engine model Version. Format: `projects/{project}/models/{model}/versions/{version}`.

Config of migrating one resource from automl.googleapis.com, datalabeling.googleapis.com and ml.googleapis.com to Vertex AI.

Used in: BatchMigrateResourcesOperationMetadata.PartialResult, BatchMigrateResourcesRequest

oneof request
- MigrateResourceRequest.MigrateMlEngineModelVersionConfig migrate_ml_engine_model_version_config = 1
  Config for migrating Version in ml.googleapis.com to Vertex AI's Model.
- MigrateResourceRequest.MigrateAutomlModelConfig migrate_automl_model_config = 2
  Config for migrating Model in automl.googleapis.com to Vertex AI's Model.
- MigrateResourceRequest.MigrateAutomlDatasetConfig migrate_automl_dataset_config = 3
  Config for migrating Dataset in automl.googleapis.com to Vertex AI's Dataset.
- MigrateResourceRequest.MigrateDataLabelingDatasetConfig migrate_data_labeling_dataset_config = 4
  Config for migrating Dataset in datalabeling.googleapis.com to Vertex AI's Dataset.

Config for migrating Dataset in automl.googleapis.com to Vertex AI's Dataset.

string dataset = 1
Required. Full resource name of automl Dataset. Format: `projects/{project}/locations/{location}/datasets/{dataset}`.
string dataset_display_name = 2
Required. Display name of the Dataset in Vertex AI. System will pick a display name if unspecified.

Config for migrating Model in automl.googleapis.com to Vertex AI's Model.

string model = 1
Required. Full resource name of automl Model. Format: `projects/{project}/locations/{location}/models/{model}`.
string model_display_name = 2
Optional. Display name of the model in Vertex AI. System will pick a display name if unspecified.

Config for migrating Dataset in datalabeling.googleapis.com to Vertex AI's Dataset.

string dataset = 1
Required. Full resource name of data labeling Dataset. Format: `projects/{project}/datasets/{dataset}`.
string dataset_display_name = 2
Optional. Display name of the Dataset in Vertex AI. System will pick a display name if unspecified.
repeated MigrateDataLabelingDatasetConfig.MigrateDataLabelingAnnotatedDatasetConfig migrate_data_labeling_annotated_dataset_configs = 3
Optional. Configs for migrating AnnotatedDataset in datalabeling.googleapis.com to Vertex AI's SavedQuery. The specified AnnotatedDatasets have to belong to the datalabeling Dataset.

Config for migrating AnnotatedDataset in datalabeling.googleapis.com to Vertex AI's SavedQuery.

Used in: MigrateDataLabelingDatasetConfig

string annotated_dataset = 1
Required. Full resource name of data labeling AnnotatedDataset. Format: `projects/{project}/datasets/{dataset}/annotatedDatasets/{annotated_dataset}`.

Config for migrating version in ml.googleapis.com to Vertex AI's Model.

string endpoint = 1
Required. The ml.googleapis.com endpoint that this model version should be migrated from. Example values: * ml.googleapis.com * us-centrall-ml.googleapis.com * europe-west4-ml.googleapis.com * asia-east1-ml.googleapis.com
string model_version = 2
Required. Full resource name of ml engine model version. Format: `projects/{project}/models/{model}/versions/{version}`.
string model_display_name = 3
Required. Display name of the model in Vertex AI. System will pick a display name if unspecified.

Describes a successfully migrated resource.

Used in: BatchMigrateResourcesResponse

oneof migrated_resource
After migration, the resource name in Vertex AI.
- string dataset = 1
  Migrated Dataset's resource name.
- string model = 2
  Migrated Model's resource name.
optional MigratableResource migratable_resource = 3
Before migration, the identifier in ml.googleapis.com, automl.googleapis.com or datalabeling.googleapis.com.

Content Part modality

Used in: ModalityTokenCount

MODALITY_UNSPECIFIED = 0
Unspecified modality.
TEXT = 1
Plain text.
IMAGE = 2
Image.
VIDEO = 3
Video.
AUDIO = 4
Audio.
DOCUMENT = 5
Document, e.g. PDF.

Represents token counting info for a single modality.

Used in: CountTokensResponse, GenerateContentResponse.UsageMetadata

Modality modality = 1
The modality associated with this token count.
int32 token_count = 2
Number of tokens.

A trained machine learning Model.

Used as response type in: ModelService.GetModel, ModelService.MergeVersionAliases, ModelService.UpdateModel

Used as field type in: ListModelVersionsResponse, ListModelsResponse, TrainingPipeline, UpdateModelRequest, UploadModelRequest

string name = 1
The resource name of the Model.
string version_id = 28
Output only. Immutable. The version ID of the model. A new version is committed when a new model version is uploaded or trained under an existing model id. It is an auto-incrementing decimal number in string representation.
repeated string version_aliases = 29
User provided version aliases so that a model version can be referenced via alias (i.e. `projects/{project}/locations/{location}/models/{model_id}@{version_alias}` instead of auto-generated version id (i.e. `projects/{project}/locations/{location}/models/{model_id}@{version_id})`. The format is [a-z][a-zA-Z0-9-]{0,126}[a-z0-9] to distinguish from version_id. A default version alias will be created for the first version of the model, and there must be exactly one default version alias for a model.
optional protobuf.Timestamp version_create_time = 31
Output only. Timestamp when this version was created.
optional protobuf.Timestamp version_update_time = 32
Output only. Timestamp when this version was most recently updated.
string display_name = 2
Required. The display name of the Model. The name can be up to 128 characters long and can consist of any UTF-8 characters.
string description = 3
The description of the Model.
string version_description = 30
The description of this version.
string default_checkpoint_id = 53
The default checkpoint id of a model version.
optional PredictSchemata predict_schemata = 4
The schemata that describe formats of the Model's predictions and explanations as given and returned via [PredictionService.Predict][google.cloud.aiplatform.v1.PredictionService.Predict] and [PredictionService.Explain][google.cloud.aiplatform.v1.PredictionService.Explain].
string metadata_schema_uri = 5
Immutable. Points to a YAML file stored on Google Cloud Storage describing additional information about the Model, that is specific to it. Unset if the Model does not have any additional information. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). AutoML Models always have this field populated by Vertex AI, if no additional metadata is needed, this field is set to an empty string. Note: The URI given on output will be immutable and probably different, including the URI scheme, than the one given on input. The output URI will point to a location where the user only has a read access.
optional protobuf.Value metadata = 6
Immutable. An additional information about the Model; the schema of the metadata can be found in [metadata_schema][google.cloud.aiplatform.v1.Model.metadata_schema_uri]. Unset if the Model does not have any additional information.
repeated Model.ExportFormat supported_export_formats = 20
Output only. The formats in which this Model may be exported. If empty, this Model is not available for export.
string training_pipeline = 7
Output only. The resource name of the TrainingPipeline that uploaded this Model, if any.
string pipeline_job = 47
Optional. This field is populated if the model is produced by a pipeline job.
optional ModelContainerSpec container_spec = 9
Input only. The specification of the container that is to be used when deploying this Model. The specification is ingested upon [ModelService.UploadModel][google.cloud.aiplatform.v1.ModelService.UploadModel], and all binaries it contains are copied and stored internally by Vertex AI. Not required for AutoML Models.
string artifact_uri = 26
Immutable. The path to the directory containing the Model artifact and any of its supporting files. Not required for AutoML Models.
repeated Model.DeploymentResourcesType supported_deployment_resources_types = 10
Output only. When this Model is deployed, its prediction resources are described by the `prediction_resources` field of the [Endpoint.deployed_models][google.cloud.aiplatform.v1.Endpoint.deployed_models] object. Because not all Models support all resource configuration types, the configuration types this Model supports are listed here. If no configuration types are listed, the Model cannot be deployed to an [Endpoint][google.cloud.aiplatform.v1.Endpoint] and does not support online predictions ([PredictionService.Predict][google.cloud.aiplatform.v1.PredictionService.Predict] or [PredictionService.Explain][google.cloud.aiplatform.v1.PredictionService.Explain]). Such a Model can serve predictions by using a [BatchPredictionJob][google.cloud.aiplatform.v1.BatchPredictionJob], if it has at least one entry each in [supported_input_storage_formats][google.cloud.aiplatform.v1.Model.supported_input_storage_formats] and [supported_output_storage_formats][google.cloud.aiplatform.v1.Model.supported_output_storage_formats].
repeated string supported_input_storage_formats = 11
Output only. The formats this Model supports in [BatchPredictionJob.input_config][google.cloud.aiplatform.v1.BatchPredictionJob.input_config]. If [PredictSchemata.instance_schema_uri][google.cloud.aiplatform.v1.PredictSchemata.instance_schema_uri] exists, the instances should be given as per that schema. The possible formats are: * `jsonl` The JSON Lines format, where each instance is a single line. Uses [GcsSource][google.cloud.aiplatform.v1.BatchPredictionJob.InputConfig.gcs_source]. * `csv` The CSV format, where each instance is a single comma-separated line. The first line in the file is the header, containing comma-separated field names. Uses [GcsSource][google.cloud.aiplatform.v1.BatchPredictionJob.InputConfig.gcs_source]. * `tf-record` The TFRecord format, where each instance is a single record in tfrecord syntax. Uses [GcsSource][google.cloud.aiplatform.v1.BatchPredictionJob.InputConfig.gcs_source]. * `tf-record-gzip` Similar to `tf-record`, but the file is gzipped. Uses [GcsSource][google.cloud.aiplatform.v1.BatchPredictionJob.InputConfig.gcs_source]. * `bigquery` Each instance is a single row in BigQuery. Uses [BigQuerySource][google.cloud.aiplatform.v1.BatchPredictionJob.InputConfig.bigquery_source]. * `file-list` Each line of the file is the location of an instance to process, uses `gcs_source` field of the [InputConfig][google.cloud.aiplatform.v1.BatchPredictionJob.InputConfig] object. If this Model doesn't support any of these formats it means it cannot be used with a [BatchPredictionJob][google.cloud.aiplatform.v1.BatchPredictionJob]. However, if it has [supported_deployment_resources_types][google.cloud.aiplatform.v1.Model.supported_deployment_resources_types], it could serve online predictions by using [PredictionService.Predict][google.cloud.aiplatform.v1.PredictionService.Predict] or [PredictionService.Explain][google.cloud.aiplatform.v1.PredictionService.Explain].
repeated string supported_output_storage_formats = 12
Output only. The formats this Model supports in [BatchPredictionJob.output_config][google.cloud.aiplatform.v1.BatchPredictionJob.output_config]. If both [PredictSchemata.instance_schema_uri][google.cloud.aiplatform.v1.PredictSchemata.instance_schema_uri] and [PredictSchemata.prediction_schema_uri][google.cloud.aiplatform.v1.PredictSchemata.prediction_schema_uri] exist, the predictions are returned together with their instances. In other words, the prediction has the original instance data first, followed by the actual prediction content (as per the schema). The possible formats are: * `jsonl` The JSON Lines format, where each prediction is a single line. Uses [GcsDestination][google.cloud.aiplatform.v1.BatchPredictionJob.OutputConfig.gcs_destination]. * `csv` The CSV format, where each prediction is a single comma-separated line. The first line in the file is the header, containing comma-separated field names. Uses [GcsDestination][google.cloud.aiplatform.v1.BatchPredictionJob.OutputConfig.gcs_destination]. * `bigquery` Each prediction is a single row in a BigQuery table, uses [BigQueryDestination][google.cloud.aiplatform.v1.BatchPredictionJob.OutputConfig.bigquery_destination] . If this Model doesn't support any of these formats it means it cannot be used with a [BatchPredictionJob][google.cloud.aiplatform.v1.BatchPredictionJob]. However, if it has [supported_deployment_resources_types][google.cloud.aiplatform.v1.Model.supported_deployment_resources_types], it could serve online predictions by using [PredictionService.Predict][google.cloud.aiplatform.v1.PredictionService.Predict] or [PredictionService.Explain][google.cloud.aiplatform.v1.PredictionService.Explain].
optional protobuf.Timestamp create_time = 13
Output only. Timestamp when this Model was uploaded into Vertex AI.
optional protobuf.Timestamp update_time = 14
Output only. Timestamp when this Model was most recently updated.
repeated DeployedModelRef deployed_models = 15
Output only. The pointers to DeployedModels created from this Model. Note that Model could have been deployed to Endpoints in different Locations.
optional ExplanationSpec explanation_spec = 23
The default explanation specification for this Model. The Model can be used for [requesting explanation][google.cloud.aiplatform.v1.PredictionService.Explain] after being [deployed][google.cloud.aiplatform.v1.EndpointService.DeployModel] if it is populated. The Model can be used for [batch explanation][google.cloud.aiplatform.v1.BatchPredictionJob.generate_explanation] if it is populated. All fields of the explanation_spec can be overridden by [explanation_spec][google.cloud.aiplatform.v1.DeployedModel.explanation_spec] of [DeployModelRequest.deployed_model][google.cloud.aiplatform.v1.DeployModelRequest.deployed_model], or [explanation_spec][google.cloud.aiplatform.v1.BatchPredictionJob.explanation_spec] of [BatchPredictionJob][google.cloud.aiplatform.v1.BatchPredictionJob]. If the default explanation specification is not set for this Model, this Model can still be used for [requesting explanation][google.cloud.aiplatform.v1.PredictionService.Explain] by setting [explanation_spec][google.cloud.aiplatform.v1.DeployedModel.explanation_spec] of [DeployModelRequest.deployed_model][google.cloud.aiplatform.v1.DeployModelRequest.deployed_model] and for [batch explanation][google.cloud.aiplatform.v1.BatchPredictionJob.generate_explanation] by setting [explanation_spec][google.cloud.aiplatform.v1.BatchPredictionJob.explanation_spec] of [BatchPredictionJob][google.cloud.aiplatform.v1.BatchPredictionJob].
string etag = 16
Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 17
The labels with user-defined metadata to organize your Models. 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.
optional Model.DataStats data_stats = 21
Stats of data used for training or evaluating the Model. Only populated when the Model is trained by a TrainingPipeline with [data_input_config][google.cloud.aiplatform.v1.TrainingPipeline.input_data_config].
optional EncryptionSpec encryption_spec = 24
Customer-managed encryption key spec for a Model. If set, this Model and all sub-resources of this Model will be secured by this key.
optional ModelSourceInfo model_source_info = 38
Output only. Source of a model. It can either be automl training pipeline, custom training pipeline, BigQuery ML, or saved and tuned from Genie or Model Garden.
optional Model.OriginalModelInfo original_model_info = 34
Output only. If this Model is a copy of another Model, this contains info about the original.
string metadata_artifact = 44
Output only. The resource name of the Artifact that was created in MetadataStore when creating the Model. The Artifact resource name pattern is `projects/{project}/locations/{location}/metadataStores/{metadata_store}/artifacts/{artifact}`.
optional Model.BaseModelSource base_model_source = 50
Optional. User input field to specify the base model source. Currently it only supports specifing the Model Garden models and Genie models.
bool satisfies_pzs = 51
Output only. Reserved for future use.
bool satisfies_pzi = 52
Output only. Reserved for future use.
repeated Checkpoint checkpoints = 57
Optional. Output only. The checkpoints of the model.

User input field to specify the base model source. Currently it only supports specifing the Model Garden models and Genie models.

Used in: Model

oneof source
- ModelGardenSource model_garden_source = 1
  Source information of Model Garden models.
- GenieSource genie_source = 2
  Information about the base model of Genie models.

Stats of data used for train or evaluate the Model.

Used in: ExportDataResponse, Model

int64 training_data_items_count = 1
Number of DataItems that were used for training this Model.
int64 validation_data_items_count = 2
Number of DataItems that were used for validating this Model during training.
int64 test_data_items_count = 3
Number of DataItems that were used for evaluating this Model. If the Model is evaluated multiple times, this will be the number of test DataItems used by the first evaluation. If the Model is not evaluated, the number is 0.
int64 training_annotations_count = 4
Number of Annotations that are used for training this Model.
int64 validation_annotations_count = 5
Number of Annotations that are used for validating this Model during training.
int64 test_annotations_count = 6
Number of Annotations that are used for evaluating this Model. If the Model is evaluated multiple times, this will be the number of test Annotations used by the first evaluation. If the Model is not evaluated, the number is 0.

Identifies a type of Model's prediction resources.

Used in: Model

DEPLOYMENT_RESOURCES_TYPE_UNSPECIFIED = 0
Should not be used.
DEDICATED_RESOURCES = 1
Resources that are dedicated to the [DeployedModel][google.cloud.aiplatform.v1.DeployedModel], and that need a higher degree of manual configuration.
AUTOMATIC_RESOURCES = 2
Resources that to large degree are decided by Vertex AI, and require only a modest additional configuration.
SHARED_RESOURCES = 3
Resources that can be shared by multiple [DeployedModels][google.cloud.aiplatform.v1.DeployedModel]. A pre-configured [DeploymentResourcePool][google.cloud.aiplatform.v1.DeploymentResourcePool] is required.

Represents export format supported by the Model. All formats export to Google Cloud Storage.

Used in: Model

string id = 1
Output only. The ID of the export format. The possible format IDs are: * `tflite` Used for Android mobile devices. * `edgetpu-tflite` Used for [Edge TPU](https://cloud.google.com/edge-tpu/) devices. * `tf-saved-model` A tensorflow model in SavedModel format. * `tf-js` A [TensorFlow.js](https://www.tensorflow.org/js) model that can be used in the browser and in Node.js using JavaScript. * `core-ml` Used for iOS mobile devices. * `custom-trained` A Model that was uploaded or trained by custom code.
repeated ExportFormat.ExportableContent exportable_contents = 2
Output only. The content of this Model that may be exported.

The Model content that can be exported.

Used in: ExportFormat

EXPORTABLE_CONTENT_UNSPECIFIED = 0
Should not be used.
ARTIFACT = 1
Model artifact and any of its supported files. Will be exported to the location specified by the `artifactDestination` field of the [ExportModelRequest.output_config][google.cloud.aiplatform.v1.ExportModelRequest.output_config] object.
IMAGE = 2
The container image that is to be used when deploying this Model. Will be exported to the location specified by the `imageDestination` field of the [ExportModelRequest.output_config][google.cloud.aiplatform.v1.ExportModelRequest.output_config] object.

Contains information about the original Model if this Model is a copy.

Used in: Model

string model = 1
Output only. The resource name of the Model this Model is a copy of, including the revision. Format: `projects/{project}/locations/{location}/models/{model_id}@{version_id}`

Specification of a container for serving predictions. Some fields in this message correspond to fields in the [Kubernetes Container v1 core specification](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.23/#container-v1-core).

Used in: Model, PublisherModel.CallToAction.Deploy, UnmanagedContainerModel

string image_uri = 1
Required. Immutable. URI of the Docker image to be used as the custom container for serving predictions. This URI must identify an image in Artifact Registry or Container Registry. Learn more about the [container publishing requirements](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#publishing), including permissions requirements for the Vertex AI Service Agent. The container image is ingested upon [ModelService.UploadModel][google.cloud.aiplatform.v1.ModelService.UploadModel], stored internally, and this original path is afterwards not used. To learn about the requirements for the Docker image itself, see [Custom container requirements](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#). You can use the URI to one of Vertex AI's [pre-built container images for prediction](https://cloud.google.com/vertex-ai/docs/predictions/pre-built-containers) in this field.
repeated string command = 2
Immutable. Specifies the command that runs when the container starts. This overrides the container's [ENTRYPOINT](https://docs.docker.com/engine/reference/builder/#entrypoint). Specify this field as an array of executable and arguments, similar to a Docker `ENTRYPOINT`'s "exec" form, not its "shell" form. If you do not specify this field, then the container's `ENTRYPOINT` runs, in conjunction with the [args][google.cloud.aiplatform.v1.ModelContainerSpec.args] field or the container's [`CMD`](https://docs.docker.com/engine/reference/builder/#cmd), if either exists. If this field is not specified and the container does not have an `ENTRYPOINT`, then refer to the Docker documentation about [how `CMD` and `ENTRYPOINT` interact](https://docs.docker.com/engine/reference/builder/#understand-how-cmd-and-entrypoint-interact). If you specify this field, then you can also specify the `args` field to provide additional arguments for this command. However, if you specify this field, then the container's `CMD` is ignored. See the [Kubernetes documentation about how the `command` and `args` fields interact with a container's `ENTRYPOINT` and `CMD`](https://kubernetes.io/docs/tasks/inject-data-application/define-command-argument-container/#notes). In this field, you can reference [environment variables set by Vertex AI](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#aip-variables) and environment variables set in the [env][google.cloud.aiplatform.v1.ModelContainerSpec.env] field. You cannot reference environment variables set in the Docker image. In order for environment variables to be expanded, reference them by using the following syntax: <code>$(<var>VARIABLE_NAME</var>)</code> Note that this differs from Bash variable expansion, which does not use parentheses. If a variable cannot be resolved, the reference in the input string is used unchanged. To avoid variable expansion, you can escape this syntax with `$$`; for example: <code>$$(<var>VARIABLE_NAME</var>)</code> This field corresponds to the `command` field of the Kubernetes Containers [v1 core API](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.23/#container-v1-core).
repeated string args = 3
Immutable. Specifies arguments for the command that runs when the container starts. This overrides the container's [`CMD`](https://docs.docker.com/engine/reference/builder/#cmd). Specify this field as an array of executable and arguments, similar to a Docker `CMD`'s "default parameters" form. If you don't specify this field but do specify the [command][google.cloud.aiplatform.v1.ModelContainerSpec.command] field, then the command from the `command` field runs without any additional arguments. See the [Kubernetes documentation about how the `command` and `args` fields interact with a container's `ENTRYPOINT` and `CMD`](https://kubernetes.io/docs/tasks/inject-data-application/define-command-argument-container/#notes). If you don't specify this field and don't specify the `command` field, then the container's [`ENTRYPOINT`](https://docs.docker.com/engine/reference/builder/#cmd) and `CMD` determine what runs based on their default behavior. See the Docker documentation about [how `CMD` and `ENTRYPOINT` interact](https://docs.docker.com/engine/reference/builder/#understand-how-cmd-and-entrypoint-interact). In this field, you can reference [environment variables set by Vertex AI](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#aip-variables) and environment variables set in the [env][google.cloud.aiplatform.v1.ModelContainerSpec.env] field. You cannot reference environment variables set in the Docker image. In order for environment variables to be expanded, reference them by using the following syntax: <code>$(<var>VARIABLE_NAME</var>)</code> Note that this differs from Bash variable expansion, which does not use parentheses. If a variable cannot be resolved, the reference in the input string is used unchanged. To avoid variable expansion, you can escape this syntax with `$$`; for example: <code>$$(<var>VARIABLE_NAME</var>)</code> This field corresponds to the `args` field of the Kubernetes Containers [v1 core API](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.23/#container-v1-core).
repeated EnvVar env = 4
Immutable. List of environment variables to set in the container. After the container starts running, code running in the container can read these environment variables. Additionally, the [command][google.cloud.aiplatform.v1.ModelContainerSpec.command] and [args][google.cloud.aiplatform.v1.ModelContainerSpec.args] fields can reference these variables. Later entries in this list can also reference earlier entries. For example, the following example sets the variable `VAR_2` to have the value `foo bar`: ```json [ { "name": "VAR_1", "value": "foo" }, { "name": "VAR_2", "value": "$(VAR_1) bar" } ] ``` If you switch the order of the variables in the example, then the expansion does not occur. This field corresponds to the `env` field of the Kubernetes Containers [v1 core API](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.23/#container-v1-core).
repeated Port ports = 5
Immutable. List of ports to expose from the container. Vertex AI sends any prediction requests that it receives to the first port on this list. Vertex AI also sends [liveness and health checks](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#liveness) to this port. If you do not specify this field, it defaults to following value: ```json [ { "containerPort": 8080 } ] ``` Vertex AI does not use ports other than the first one listed. This field corresponds to the `ports` field of the Kubernetes Containers [v1 core API](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.23/#container-v1-core).
string predict_route = 6
Immutable. HTTP path on the container to send prediction requests to. Vertex AI forwards requests sent using [projects.locations.endpoints.predict][google.cloud.aiplatform.v1.PredictionService.Predict] to this path on the container's IP address and port. Vertex AI then returns the container's response in the API response. For example, if you set this field to `/foo`, then when Vertex AI receives a prediction request, it forwards the request body in a POST request to the `/foo` path on the port of your container specified by the first value of this `ModelContainerSpec`'s [ports][google.cloud.aiplatform.v1.ModelContainerSpec.ports] field. If you don't specify this field, it defaults to the following value when you [deploy this Model to an Endpoint][google.cloud.aiplatform.v1.EndpointService.DeployModel]: <code>/v1/endpoints/<var>ENDPOINT</var>/deployedModels/<var>DEPLOYED_MODEL</var>:predict</code> The placeholders in this value are replaced as follows: * <var>ENDPOINT</var>: The last segment (following `endpoints/`)of the Endpoint.name][] field of the Endpoint where this Model has been deployed. (Vertex AI makes this value available to your container code as the [`AIP_ENDPOINT_ID` environment variable](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#aip-variables).) * <var>DEPLOYED_MODEL</var>: [DeployedModel.id][google.cloud.aiplatform.v1.DeployedModel.id] of the `DeployedModel`. (Vertex AI makes this value available to your container code as the [`AIP_DEPLOYED_MODEL_ID` environment variable](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#aip-variables).)
string health_route = 7
Immutable. HTTP path on the container to send health checks to. Vertex AI intermittently sends GET requests to this path on the container's IP address and port to check that the container is healthy. Read more about [health checks](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#health). For example, if you set this field to `/bar`, then Vertex AI intermittently sends a GET request to the `/bar` path on the port of your container specified by the first value of this `ModelContainerSpec`'s [ports][google.cloud.aiplatform.v1.ModelContainerSpec.ports] field. If you don't specify this field, it defaults to the following value when you [deploy this Model to an Endpoint][google.cloud.aiplatform.v1.EndpointService.DeployModel]: <code>/v1/endpoints/<var>ENDPOINT</var>/deployedModels/<var>DEPLOYED_MODEL</var>:predict</code> The placeholders in this value are replaced as follows: * <var>ENDPOINT</var>: The last segment (following `endpoints/`)of the Endpoint.name][] field of the Endpoint where this Model has been deployed. (Vertex AI makes this value available to your container code as the [`AIP_ENDPOINT_ID` environment variable](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#aip-variables).) * <var>DEPLOYED_MODEL</var>: [DeployedModel.id][google.cloud.aiplatform.v1.DeployedModel.id] of the `DeployedModel`. (Vertex AI makes this value available to your container code as the [`AIP_DEPLOYED_MODEL_ID` environment variable](https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#aip-variables).)
repeated Port grpc_ports = 9
Immutable. List of ports to expose from the container. Vertex AI sends gRPC prediction requests that it receives to the first port on this list. Vertex AI also sends liveness and health checks to this port. If you do not specify this field, gRPC requests to the container will be disabled. Vertex AI does not use ports other than the first one listed. This field corresponds to the `ports` field of the Kubernetes Containers v1 core API.
optional protobuf.Duration deployment_timeout = 10
Immutable. Deployment timeout. Limit for deployment timeout is 2 hours.
int64 shared_memory_size_mb = 11
Immutable. The amount of the VM memory to reserve as the shared memory for the model in megabytes.
optional Probe startup_probe = 12
Immutable. Specification for Kubernetes startup probe.
optional Probe health_probe = 13
Immutable. Specification for Kubernetes readiness probe.
optional Probe liveness_probe = 14
Immutable. Specification for Kubernetes liveness probe.

ModelDeploymentMonitoringBigQueryTable specifies the BigQuery table name as well as some information of the logs stored in this table.

ModelDeploymentMonitoringBigQueryTable.LogSource log_source = 1
The source of log.
ModelDeploymentMonitoringBigQueryTable.LogType log_type = 2
The type of log.
string bigquery_table_path = 3
The created BigQuery table to store logs. Customer could do their own query & analysis. Format: `bq://<project_id>.model_deployment_monitoring_<endpoint_id>.<tolower(log_source)>_<tolower(log_type)>`
string request_response_logging_schema_version = 4
Output only. The schema version of the request/response logging BigQuery table. Default to v1 if unset.

Indicates where does the log come from.

Used in: ModelDeploymentMonitoringBigQueryTable

LOG_SOURCE_UNSPECIFIED = 0
Unspecified source.
TRAINING = 1
Logs coming from Training dataset.
SERVING = 2
Logs coming from Serving traffic.

Indicates what type of traffic does the log belong to.

Used in: ModelDeploymentMonitoringBigQueryTable

LOG_TYPE_UNSPECIFIED = 0
Unspecified type.
PREDICT = 1
Predict logs.
EXPLAIN = 2
Explain logs.

Represents a job that runs periodically to monitor the deployed models in an endpoint. It will analyze the logged training & prediction data to detect any abnormal behaviors.

Used as response type in: JobService.CreateModelDeploymentMonitoringJob, JobService.GetModelDeploymentMonitoringJob

Used as field type in: CreateModelDeploymentMonitoringJobRequest, ListModelDeploymentMonitoringJobsResponse, UpdateModelDeploymentMonitoringJobRequest

string name = 1
Output only. Resource name of a ModelDeploymentMonitoringJob.
string display_name = 2
Required. The user-defined name of the ModelDeploymentMonitoringJob. The name can be up to 128 characters long and can consist of any UTF-8 characters. Display name of a ModelDeploymentMonitoringJob.
string endpoint = 3
Required. Endpoint resource name. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
JobState state = 4
Output only. The detailed state of the monitoring job. When the job is still creating, the state will be 'PENDING'. Once the job is successfully created, the state will be 'RUNNING'. Pause the job, the state will be 'PAUSED'. Resume the job, the state will return to 'RUNNING'.
ModelDeploymentMonitoringJob.MonitoringScheduleState schedule_state = 5
Output only. Schedule state when the monitoring job is in Running state.
optional ModelDeploymentMonitoringJob.LatestMonitoringPipelineMetadata latest_monitoring_pipeline_metadata = 25
Output only. Latest triggered monitoring pipeline metadata.
repeated ModelDeploymentMonitoringObjectiveConfig model_deployment_monitoring_objective_configs = 6
Required. The config for monitoring objectives. This is a per DeployedModel config. Each DeployedModel needs to be configured separately.
optional ModelDeploymentMonitoringScheduleConfig model_deployment_monitoring_schedule_config = 7
Required. Schedule config for running the monitoring job.
optional SamplingStrategy logging_sampling_strategy = 8
Required. Sample Strategy for logging.
optional ModelMonitoringAlertConfig model_monitoring_alert_config = 15
Alert config for model monitoring.
string predict_instance_schema_uri = 9
YAML schema file uri describing the format of a single instance, which are given to format this Endpoint's prediction (and explanation). If not set, we will generate predict schema from collected predict requests.
optional protobuf.Value sample_predict_instance = 19
Sample Predict instance, same format as [PredictRequest.instances][google.cloud.aiplatform.v1.PredictRequest.instances], this can be set as a replacement of [ModelDeploymentMonitoringJob.predict_instance_schema_uri][google.cloud.aiplatform.v1.ModelDeploymentMonitoringJob.predict_instance_schema_uri]. If not set, we will generate predict schema from collected predict requests.
string analysis_instance_schema_uri = 16
YAML schema file uri describing the format of a single instance that you want Tensorflow Data Validation (TFDV) to analyze. If this field is empty, all the feature data types are inferred from [predict_instance_schema_uri][google.cloud.aiplatform.v1.ModelDeploymentMonitoringJob.predict_instance_schema_uri], meaning that TFDV will use the data in the exact format(data type) as prediction request/response. If there are any data type differences between predict instance and TFDV instance, this field can be used to override the schema. For models trained with Vertex AI, this field must be set as all the fields in predict instance formatted as string.
repeated ModelDeploymentMonitoringBigQueryTable bigquery_tables = 10
Output only. The created bigquery tables for the job under customer project. Customer could do their own query & analysis. There could be 4 log tables in maximum: 1. Training data logging predict request/response 2. Serving data logging predict request/response
optional protobuf.Duration log_ttl = 17
The TTL of BigQuery tables in user projects which stores logs. A day is the basic unit of the TTL and we take the ceil of TTL/86400(a day). e.g. { second: 3600} indicates ttl = 1 day.
map<string, string> labels = 11
The labels with user-defined metadata to organize your ModelDeploymentMonitoringJob. 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.
optional protobuf.Timestamp create_time = 12
Output only. Timestamp when this ModelDeploymentMonitoringJob was created.
optional protobuf.Timestamp update_time = 13
Output only. Timestamp when this ModelDeploymentMonitoringJob was updated most recently.
optional protobuf.Timestamp next_schedule_time = 14
Output only. Timestamp when this monitoring pipeline will be scheduled to run for the next round.
optional GcsDestination stats_anomalies_base_directory = 20
Stats anomalies base folder path.
optional EncryptionSpec encryption_spec = 21
Customer-managed encryption key spec for a ModelDeploymentMonitoringJob. If set, this ModelDeploymentMonitoringJob and all sub-resources of this ModelDeploymentMonitoringJob will be secured by this key.
bool enable_monitoring_pipeline_logs = 22
If true, the scheduled monitoring pipeline logs are sent to Google Cloud Logging, including pipeline status and anomalies detected. Please note the logs incur cost, which are subject to [Cloud Logging pricing](https://cloud.google.com/logging#pricing).
optional rpc.Status error = 23
Output only. Only populated when the job's state is `JOB_STATE_FAILED` or `JOB_STATE_CANCELLED`.
bool satisfies_pzs = 26
Output only. Reserved for future use.
bool satisfies_pzi = 27
Output only. Reserved for future use.

All metadata of most recent monitoring pipelines.

optional protobuf.Timestamp run_time = 1
The time that most recent monitoring pipelines that is related to this run.
optional rpc.Status status = 2
The status of the most recent monitoring pipeline.

The state to Specify the monitoring pipeline.

MONITORING_SCHEDULE_STATE_UNSPECIFIED = 0
Unspecified state.
PENDING = 1
The pipeline is picked up and wait to run.
OFFLINE = 2
The pipeline is offline and will be scheduled for next run.
RUNNING = 3
The pipeline is running.

ModelDeploymentMonitoringObjectiveConfig contains the pair of deployed_model_id to ModelMonitoringObjectiveConfig.

Used in: ModelMonitoringStatsAnomalies, SearchModelDeploymentMonitoringStatsAnomaliesRequest.StatsAnomaliesObjective

string deployed_model_id = 1
The DeployedModel ID of the objective config.
optional ModelMonitoringObjectiveConfig objective_config = 2
The objective config of for the modelmonitoring job of this deployed model.

The Model Monitoring Objective types.

MODEL_DEPLOYMENT_MONITORING_OBJECTIVE_TYPE_UNSPECIFIED = 0
Default value, should not be set.
RAW_FEATURE_SKEW = 1
Raw feature values' stats to detect skew between Training-Prediction datasets.
RAW_FEATURE_DRIFT = 2
Raw feature values' stats to detect drift between Serving-Prediction datasets.
FEATURE_ATTRIBUTION_SKEW = 3
Feature attribution scores to detect skew between Training-Prediction datasets.
FEATURE_ATTRIBUTION_DRIFT = 4
Feature attribution scores to detect skew between Prediction datasets collected within different time windows.

The config for scheduling monitoring job.

optional protobuf.Duration monitor_interval = 1
Required. The model monitoring job scheduling interval. It will be rounded up to next full hour. This defines how often the monitoring jobs are triggered.
optional protobuf.Duration monitor_window = 2
The time window of the prediction data being included in each prediction dataset. This window specifies how long the data should be collected from historical model results for each run. If not set, [ModelDeploymentMonitoringScheduleConfig.monitor_interval][google.cloud.aiplatform.v1.ModelDeploymentMonitoringScheduleConfig.monitor_interval] will be used. e.g. If currently the cutoff time is 2022-01-08 14:30:00 and the monitor_window is set to be 3600, then data from 2022-01-08 13:30:00 to 2022-01-08 14:30:00 will be retrieved and aggregated to calculate the monitoring statistics.

A collection of metrics calculated by comparing Model's predictions on all of the test data against annotations from the test data.

Used as response type in: ModelService.GetModelEvaluation, ModelService.ImportModelEvaluation

Used as field type in: ImportModelEvaluationRequest, ListModelEvaluationsResponse

string name = 1
Output only. The resource name of the ModelEvaluation.
string display_name = 10
The display name of the ModelEvaluation.
string metrics_schema_uri = 2
Points to a YAML file stored on Google Cloud Storage describing the [metrics][google.cloud.aiplatform.v1.ModelEvaluation.metrics] of this ModelEvaluation. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject).
optional protobuf.Value metrics = 3
Evaluation metrics of the Model. The schema of the metrics is stored in [metrics_schema_uri][google.cloud.aiplatform.v1.ModelEvaluation.metrics_schema_uri]
optional protobuf.Timestamp create_time = 4
Output only. Timestamp when this ModelEvaluation was created.
repeated string slice_dimensions = 5
All possible [dimensions][google.cloud.aiplatform.v1.ModelEvaluationSlice.Slice.dimension] of ModelEvaluationSlices. The dimensions can be used as the filter of the [ModelService.ListModelEvaluationSlices][google.cloud.aiplatform.v1.ModelService.ListModelEvaluationSlices] request, in the form of `slice.dimension = <dimension>`.
string data_item_schema_uri = 6
Points to a YAML file stored on Google Cloud Storage describing [EvaluatedDataItemView.data_item_payload][] and [EvaluatedAnnotation.data_item_payload][google.cloud.aiplatform.v1.EvaluatedAnnotation.data_item_payload]. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). This field is not populated if there are neither EvaluatedDataItemViews nor EvaluatedAnnotations under this ModelEvaluation.
string annotation_schema_uri = 7
Points to a YAML file stored on Google Cloud Storage describing [EvaluatedDataItemView.predictions][], [EvaluatedDataItemView.ground_truths][], [EvaluatedAnnotation.predictions][google.cloud.aiplatform.v1.EvaluatedAnnotation.predictions], and [EvaluatedAnnotation.ground_truths][google.cloud.aiplatform.v1.EvaluatedAnnotation.ground_truths]. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). This field is not populated if there are neither EvaluatedDataItemViews nor EvaluatedAnnotations under this ModelEvaluation.
optional ModelExplanation model_explanation = 8
Aggregated explanation metrics for the Model's prediction output over the data this ModelEvaluation uses. This field is populated only if the Model is evaluated with explanations, and only for AutoML tabular Models.
repeated ModelEvaluation.ModelEvaluationExplanationSpec explanation_specs = 9
Describes the values of [ExplanationSpec][google.cloud.aiplatform.v1.ExplanationSpec] that are used for explaining the predicted values on the evaluated data.
optional protobuf.Value metadata = 11
The metadata of the ModelEvaluation. For the ModelEvaluation uploaded from Managed Pipeline, metadata contains a structured value with keys of "pipeline_job_id", "evaluation_dataset_type", "evaluation_dataset_path", "row_based_metrics_path".

Used in: ModelEvaluation

string explanation_type = 1
Explanation type. For AutoML Image Classification models, possible values are: * `image-integrated-gradients` * `image-xrai`
optional ExplanationSpec explanation_spec = 2
Explanation spec details.

A collection of metrics calculated by comparing Model's predictions on a slice of the test data against ground truth annotations.

Used as response type in: ModelService.GetModelEvaluationSlice

Used as field type in: BatchImportModelEvaluationSlicesRequest, ListModelEvaluationSlicesResponse

string name = 1
Output only. The resource name of the ModelEvaluationSlice.
optional ModelEvaluationSlice.Slice slice = 2
Output only. The slice of the test data that is used to evaluate the Model.
string metrics_schema_uri = 3
Output only. Points to a YAML file stored on Google Cloud Storage describing the [metrics][google.cloud.aiplatform.v1.ModelEvaluationSlice.metrics] of this ModelEvaluationSlice. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject).
optional protobuf.Value metrics = 4
Output only. Sliced evaluation metrics of the Model. The schema of the metrics is stored in [metrics_schema_uri][google.cloud.aiplatform.v1.ModelEvaluationSlice.metrics_schema_uri]
optional protobuf.Timestamp create_time = 5
Output only. Timestamp when this ModelEvaluationSlice was created.
optional ModelExplanation model_explanation = 6
Output only. Aggregated explanation metrics for the Model's prediction output over the data this ModelEvaluation uses. This field is populated only if the Model is evaluated with explanations, and only for tabular Models.

Definition of a slice.

Used in: ModelEvaluationSlice

string dimension = 1
Output only. The dimension of the slice. Well-known dimensions are: * `annotationSpec`: This slice is on the test data that has either ground truth or prediction with [AnnotationSpec.display_name][google.cloud.aiplatform.v1.AnnotationSpec.display_name] equals to [value][google.cloud.aiplatform.v1.ModelEvaluationSlice.Slice.value]. * `slice`: This slice is a user customized slice defined by its SliceSpec.
string value = 2
Output only. The value of the dimension in this slice.
optional Slice.SliceSpec slice_spec = 3
Output only. Specification for how the data was sliced.

Specification for how the data should be sliced.

Used in: Slice

map<string, SliceSpec.SliceConfig> configs = 1
Mapping configuration for this SliceSpec. The key is the name of the feature. By default, the key will be prefixed by "instance" as a dictionary prefix for Vertex Batch Predictions output format.

A range of values for slice(s). `low` is inclusive, `high` is exclusive.

Used in: SliceConfig

float low = 1
Inclusive low value for the range.
float high = 2
Exclusive high value for the range.

Specification message containing the config for this SliceSpec. When `kind` is selected as `value` and/or `range`, only a single slice will be computed. When `all_values` is present, a separate slice will be computed for each possible label/value for the corresponding key in `config`. Examples, with feature zip_code with values 12345, 23334, 88888 and feature country with values "US", "Canada", "Mexico" in the dataset: Example 1: { "zip_code": { "value": { "float_value": 12345.0 } } } A single slice for any data with zip_code 12345 in the dataset. Example 2: { "zip_code": { "range": { "low": 12345, "high": 20000 } } } A single slice containing data where the zip_codes between 12345 and 20000 For this example, data with the zip_code of 12345 will be in this slice. Example 3: { "zip_code": { "range": { "low": 10000, "high": 20000 } }, "country": { "value": { "string_value": "US" } } } A single slice containing data where the zip_codes between 10000 and 20000 has the country "US". For this example, data with the zip_code of 12345 and country "US" will be in this slice. Example 4: { "country": {"all_values": { "value": true } } } Three slices are computed, one for each unique country in the dataset. Example 5: { "country": { "all_values": { "value": true } }, "zip_code": { "value": { "float_value": 12345.0 } } } Three slices are computed, one for each unique country in the dataset where the zip_code is also 12345. For this example, data with zip_code 12345 and country "US" will be in one slice, zip_code 12345 and country "Canada" in another slice, and zip_code 12345 and country "Mexico" in another slice, totaling 3 slices.

Used in: SliceSpec

oneof kind
- Value value = 1
  A unique specific value for a given feature. Example: `{ "value": { "string_value": "12345" } }`
- Range range = 2
  A range of values for a numerical feature. Example: `{"range":{"low":10000.0,"high":50000.0}}` will capture 12345 and 23334 in the slice.
- protobuf.BoolValue all_values = 3
  If all_values is set to true, then all possible labels of the keyed feature will have another slice computed. Example: `{"all_values":{"value":true}}`

Single value that supports strings and floats.

Used in: SliceConfig

oneof kind
- string string_value = 1
  String type.
- float float_value = 2
  Float type.

Aggregated explanation metrics for a Model over a set of instances.

Used in: ModelEvaluation, ModelEvaluationSlice

repeated Attribution mean_attributions = 1
Output only. Aggregated attributions explaining the Model's prediction outputs over the set of instances. The attributions are grouped by outputs. For Models that predict only one output, such as regression Models that predict only one score, there is only one attibution that explains the predicted output. For Models that predict multiple outputs, such as multiclass Models that predict multiple classes, each element explains one specific item. [Attribution.output_index][google.cloud.aiplatform.v1.Attribution.output_index] can be used to identify which output this attribution is explaining. The [baselineOutputValue][google.cloud.aiplatform.v1.Attribution.baseline_output_value], [instanceOutputValue][google.cloud.aiplatform.v1.Attribution.instance_output_value] and [featureAttributions][google.cloud.aiplatform.v1.Attribution.feature_attributions] fields are averaged over the test data. NOTE: Currently AutoML tabular classification Models produce only one attribution, which averages attributions over all the classes it predicts. [Attribution.approximation_error][google.cloud.aiplatform.v1.Attribution.approximation_error] is not populated.

Contains information about the source of the models generated from Model Garden.

Used in: Model.BaseModelSource

string public_model_name = 1
Required. The model garden source model resource name.
string version_id = 3
Optional. The model garden source model version ID.
bool skip_hf_model_cache = 4
Optional. Whether to avoid pulling the model from the HF cache.

The alert config for model monitoring.

oneof alert
- ModelMonitoringAlertConfig.EmailAlertConfig email_alert_config = 1
  Email alert config.
bool enable_logging = 2
Dump the anomalies to Cloud Logging. The anomalies will be put to json payload encoded from proto [ModelMonitoringStatsAnomalies][google.cloud.aiplatform.v1.ModelMonitoringStatsAnomalies]. This can be further synced to Pub/Sub or any other services supported by Cloud Logging.
repeated string notification_channels = 3
Resource names of the NotificationChannels to send alert. Must be of the format `projects/<project_id_or_number>/notificationChannels/<channel_id>`

The config for email alert.

Used in: ModelMonitoringAlertConfig

repeated string user_emails = 1
The email addresses to send the alert.

The objective configuration for model monitoring, including the information needed to detect anomalies for one particular model.

Used in: ModelDeploymentMonitoringObjectiveConfig

optional ModelMonitoringObjectiveConfig.TrainingDataset training_dataset = 1
Training dataset for models. This field has to be set only if TrainingPredictionSkewDetectionConfig is specified.
optional ModelMonitoringObjectiveConfig.TrainingPredictionSkewDetectionConfig training_prediction_skew_detection_config = 2
The config for skew between training data and prediction data.
optional ModelMonitoringObjectiveConfig.PredictionDriftDetectionConfig prediction_drift_detection_config = 3
The config for drift of prediction data.
optional ModelMonitoringObjectiveConfig.ExplanationConfig explanation_config = 5
The config for integrating with Vertex Explainable AI.

The config for integrating with Vertex Explainable AI. Only applicable if the Model has explanation_spec populated.

bool enable_feature_attributes = 1
If want to analyze the Vertex Explainable AI feature attribute scores or not. If set to true, Vertex AI will log the feature attributions from explain response and do the skew/drift detection for them.
optional ExplanationConfig.ExplanationBaseline explanation_baseline = 2
Predictions generated by the BatchPredictionJob using baseline dataset.

Output from [BatchPredictionJob][google.cloud.aiplatform.v1.BatchPredictionJob] for Model Monitoring baseline dataset, which can be used to generate baseline attribution scores.

Used in: ExplanationConfig

oneof destination
The configuration specifying of BatchExplain job output. This can be used to generate the baseline of feature attribution scores.
- GcsDestination gcs = 2
  Cloud Storage location for BatchExplain output.
- BigQueryDestination bigquery = 3
  BigQuery location for BatchExplain output.
ExplanationBaseline.PredictionFormat prediction_format = 1
The storage format of the predictions generated BatchPrediction job.

The storage format of the predictions generated BatchPrediction job.

Used in: ExplanationBaseline

PREDICTION_FORMAT_UNSPECIFIED = 0
Should not be set.
JSONL = 2
Predictions are in JSONL files.
BIGQUERY = 3
Predictions are in BigQuery.

The config for Prediction data drift detection.

map<string, ThresholdConfig> drift_thresholds = 1
Key is the feature name and value is the threshold. If a feature needs to be monitored for drift, a value threshold must be configured for that feature. The threshold here is against feature distribution distance between different time windws.
map<string, ThresholdConfig> attribution_score_drift_thresholds = 2
Key is the feature name and value is the threshold. The threshold here is against attribution score distance between different time windows.
optional ThresholdConfig default_drift_threshold = 5
Drift anomaly detection threshold used by all features. When the per-feature thresholds are not set, this field can be used to specify a threshold for all features.

Training Dataset information.

oneof data_source
- string dataset = 3
  The resource name of the Dataset used to train this Model.
- GcsSource gcs_source = 4
  The Google Cloud Storage uri of the unmanaged Dataset used to train this Model.
- BigQuerySource bigquery_source = 5
  The BigQuery table of the unmanaged Dataset used to train this Model.
string data_format = 2
Data format of the dataset, only applicable if the input is from Google Cloud Storage. The possible formats are: "tf-record" The source file is a TFRecord file. "csv" The source file is a CSV file. "jsonl" The source file is a JSONL file.
string target_field = 6
The target field name the model is to predict. This field will be excluded when doing Predict and (or) Explain for the training data.
optional SamplingStrategy logging_sampling_strategy = 7
Strategy to sample data from Training Dataset. If not set, we process the whole dataset.

The config for Training & Prediction data skew detection. It specifies the training dataset sources and the skew detection parameters.

map<string, ThresholdConfig> skew_thresholds = 1
Key is the feature name and value is the threshold. If a feature needs to be monitored for skew, a value threshold must be configured for that feature. The threshold here is against feature distribution distance between the training and prediction feature.
map<string, ThresholdConfig> attribution_score_skew_thresholds = 2
Key is the feature name and value is the threshold. The threshold here is against attribution score distance between the training and prediction feature.
optional ThresholdConfig default_skew_threshold = 6
Skew anomaly detection threshold used by all features. When the per-feature thresholds are not set, this field can be used to specify a threshold for all features.

Statistics and anomalies generated by Model Monitoring.

Used in: SearchModelDeploymentMonitoringStatsAnomaliesResponse

ModelDeploymentMonitoringObjectiveType objective = 1
Model Monitoring Objective those stats and anomalies belonging to.
string deployed_model_id = 2
Deployed Model ID.
int32 anomaly_count = 3
Number of anomalies within all stats.
repeated ModelMonitoringStatsAnomalies.FeatureHistoricStatsAnomalies feature_stats = 4
A list of historical Stats and Anomalies generated for all Features.

Historical Stats (and Anomalies) for a specific Feature.

Used in: ModelMonitoringStatsAnomalies

string feature_display_name = 1
Display Name of the Feature.
optional ThresholdConfig threshold = 3
Threshold for anomaly detection.
optional FeatureStatsAnomaly training_stats = 4
Stats calculated for the Training Dataset.
repeated FeatureStatsAnomaly prediction_stats = 5
A list of historical stats generated by different time window's Prediction Dataset.

Detail description of the source information of the model.

Used in: Model

ModelSourceInfo.ModelSourceType source_type = 1
Type of the model source.
bool copy = 2
If this Model is copy of another Model. If true then [source_type][google.cloud.aiplatform.v1.ModelSourceInfo.source_type] pertains to the original.

Source of the model. Different from `objective` field, this `ModelSourceType` enum indicates the source from which the model was accessed or obtained, whereas the `objective` indicates the overall aim or function of this model.

Used in: ModelSourceInfo

MODEL_SOURCE_TYPE_UNSPECIFIED = 0
Should not be used.
AUTOML = 1
The Model is uploaded by automl training pipeline.
CUSTOM = 2
The Model is uploaded by user or custom training pipeline.
BQML = 3
The Model is registered and sync'ed from BigQuery ML.
MODEL_GARDEN = 4
The Model is saved or tuned from Model Garden.
GENIE = 5
The Model is saved or tuned from Genie.
CUSTOM_TEXT_EMBEDDING = 6
The Model is uploaded by text embedding finetuning pipeline.
MARKETPLACE = 7
The Model is saved or tuned from Marketplace.

A proto representation of a Spanner-stored ModelVersionCheckpoint. The meaning of the fields is equivalent to their in-Spanner counterparts.

Used in: ListModelVersionCheckpointsResponse

string checkpoint_id = 1
The ID of the checkpoint.
int64 epoch = 2
The epoch of the checkpoint.
int64 step = 3
The step of the checkpoint.

Runtime operation information for [IndexEndpointService.MutateDeployedIndex][google.cloud.aiplatform.v1.IndexEndpointService.MutateDeployedIndex].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.
string deployed_index_id = 2
The unique index id specified by user

Response message for [IndexEndpointService.MutateDeployedIndex][google.cloud.aiplatform.v1.IndexEndpointService.MutateDeployedIndex].

optional DeployedIndex deployed_index = 1
The DeployedIndex that had been updated in the IndexEndpoint.

Runtime operation information for [EndpointService.MutateDeployedModel][google.cloud.aiplatform.v1.EndpointService.MutateDeployedModel].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Response message for [EndpointService.MutateDeployedModel][google.cloud.aiplatform.v1.EndpointService.MutateDeployedModel].

optional DeployedModel deployed_model = 1
The DeployedModel that's being mutated.

Represents a Neural Architecture Search (NAS) job.

Used as response type in: JobService.CreateNasJob, JobService.GetNasJob

Used as field type in: CreateNasJobRequest, ListNasJobsResponse

string name = 1
Output only. Resource name of the NasJob.
string display_name = 2
Required. The display name of the NasJob. The name can be up to 128 characters long and can consist of any UTF-8 characters.
optional NasJobSpec nas_job_spec = 4
Required. The specification of a NasJob.
optional NasJobOutput nas_job_output = 5
Output only. Output of the NasJob.
JobState state = 6
Output only. The detailed state of the job.
optional protobuf.Timestamp create_time = 7
Output only. Time when the NasJob was created.
optional protobuf.Timestamp start_time = 8
Output only. Time when the NasJob for the first time entered the `JOB_STATE_RUNNING` state.
optional protobuf.Timestamp end_time = 9
Output only. Time when the NasJob entered any of the following states: `JOB_STATE_SUCCEEDED`, `JOB_STATE_FAILED`, `JOB_STATE_CANCELLED`.
optional protobuf.Timestamp update_time = 10
Output only. Time when the NasJob was most recently updated.
optional rpc.Status error = 11
Output only. Only populated when job's state is JOB_STATE_FAILED or JOB_STATE_CANCELLED.
map<string, string> labels = 12
The labels with user-defined metadata to organize NasJobs. 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.
optional EncryptionSpec encryption_spec = 13
Customer-managed encryption key options for a NasJob. If this is set, then all resources created by the NasJob will be encrypted with the provided encryption key.
bool enable_restricted_image_training = 14
Optional. Enable a separation of Custom model training and restricted image training for tenant project.
bool satisfies_pzs = 15
Output only. Reserved for future use.
bool satisfies_pzi = 16
Output only. Reserved for future use.

Represents a uCAIP NasJob output.

Used in: NasJob

oneof output
The output of this Neural Architecture Search (NAS) job.
- NasJobOutput.MultiTrialJobOutput multi_trial_job_output = 1
  Output only. The output of this multi-trial Neural Architecture Search (NAS) job.

The output of a multi-trial Neural Architecture Search (NAS) jobs.

Used in: NasJobOutput

repeated NasTrial search_trials = 1
Output only. List of NasTrials that were started as part of search stage.
repeated NasTrial train_trials = 2
Output only. List of NasTrials that were started as part of train stage.

Represents the spec of a NasJob.

Used in: NasJob

oneof nas_algorithm_spec
The Neural Architecture Search (NAS) algorithm specification.
- NasJobSpec.MultiTrialAlgorithmSpec multi_trial_algorithm_spec = 2
  The spec of multi-trial algorithms.
string resume_nas_job_id = 3
The ID of the existing NasJob in the same Project and Location which will be used to resume search. search_space_spec and nas_algorithm_spec are obtained from previous NasJob hence should not provide them again for this NasJob.
string search_space_spec = 1
It defines the search space for Neural Architecture Search (NAS).

The spec of multi-trial Neural Architecture Search (NAS).

Used in: NasJobSpec

MultiTrialAlgorithmSpec.MultiTrialAlgorithm multi_trial_algorithm = 1
The multi-trial Neural Architecture Search (NAS) algorithm type. Defaults to `REINFORCEMENT_LEARNING`.
optional MultiTrialAlgorithmSpec.MetricSpec metric = 2
Metric specs for the NAS job. Validation for this field is done at `multi_trial_algorithm_spec` field.
optional MultiTrialAlgorithmSpec.SearchTrialSpec search_trial_spec = 3
Required. Spec for search trials.
optional MultiTrialAlgorithmSpec.TrainTrialSpec train_trial_spec = 4
Spec for train trials. Top N [TrainTrialSpec.max_parallel_trial_count] search trials will be trained for every M [TrainTrialSpec.frequency] trials searched.

Represents a metric to optimize.

string metric_id = 1
Required. The ID of the metric. Must not contain whitespaces.
MetricSpec.GoalType goal = 2
Required. The optimization goal of the metric.

The available types of optimization goals.

Used in: MetricSpec

GOAL_TYPE_UNSPECIFIED = 0
Goal Type will default to maximize.
MAXIMIZE = 1
Maximize the goal metric.
MINIMIZE = 2
Minimize the goal metric.

The available types of multi-trial algorithms.

MULTI_TRIAL_ALGORITHM_UNSPECIFIED = 0
Defaults to `REINFORCEMENT_LEARNING`.
REINFORCEMENT_LEARNING = 1
The Reinforcement Learning Algorithm for Multi-trial Neural Architecture Search (NAS).
GRID_SEARCH = 2
The Grid Search Algorithm for Multi-trial Neural Architecture Search (NAS).

Represent spec for search trials.

optional CustomJobSpec search_trial_job_spec = 1
Required. The spec of a search trial job. The same spec applies to all search trials.
int32 max_trial_count = 2
Required. The maximum number of Neural Architecture Search (NAS) trials to run.
int32 max_parallel_trial_count = 3
Required. The maximum number of trials to run in parallel.
int32 max_failed_trial_count = 4
The number of failed trials that need to be seen before failing the NasJob. If set to 0, Vertex AI decides how many trials must fail before the whole job fails.

Represent spec for train trials.

optional CustomJobSpec train_trial_job_spec = 1
Required. The spec of a train trial job. The same spec applies to all train trials.
int32 max_parallel_trial_count = 2
Required. The maximum number of trials to run in parallel.
int32 frequency = 3
Required. Frequency of search trials to start train stage. Top N [TrainTrialSpec.max_parallel_trial_count] search trials will be trained for every M [TrainTrialSpec.frequency] trials searched.

Represents a uCAIP NasJob trial.

Used in: NasJobOutput.MultiTrialJobOutput, NasTrialDetail

string id = 1
Output only. The identifier of the NasTrial assigned by the service.
NasTrial.State state = 2
Output only. The detailed state of the NasTrial.
optional Measurement final_measurement = 3
Output only. The final measurement containing the objective value.
optional protobuf.Timestamp start_time = 4
Output only. Time when the NasTrial was started.
optional protobuf.Timestamp end_time = 5
Output only. Time when the NasTrial's status changed to `SUCCEEDED` or `INFEASIBLE`.

Describes a NasTrial state.

Used in: NasTrial

STATE_UNSPECIFIED = 0
The NasTrial state is unspecified.
REQUESTED = 1
Indicates that a specific NasTrial has been requested, but it has not yet been suggested by the service.
ACTIVE = 2
Indicates that the NasTrial has been suggested.
STOPPING = 3
Indicates that the NasTrial should stop according to the service.
SUCCEEDED = 4
Indicates that the NasTrial is completed successfully.
INFEASIBLE = 5
Indicates that the NasTrial should not be attempted again. The service will set a NasTrial to INFEASIBLE when it's done but missing the final_measurement.

Represents a NasTrial details along with its parameters. If there is a corresponding train NasTrial, the train NasTrial is also returned.

Used as response type in: JobService.GetNasTrialDetail

Used as field type in: ListNasTrialDetailsResponse

string name = 1
Output only. Resource name of the NasTrialDetail.
string parameters = 2
The parameters for the NasJob NasTrial.
optional NasTrial search_trial = 3
The requested search NasTrial.
optional NasTrial train_trial = 4
The train NasTrial corresponding to [search_trial][google.cloud.aiplatform.v1.NasTrialDetail.search_trial]. Only populated if [search_trial][google.cloud.aiplatform.v1.NasTrialDetail.search_trial] is used for training.

A query to find a number of similar entities.

Used in: SearchNearestEntitiesRequest

oneof instance
- string entity_id = 1
  Optional. The entity id whose similar entities should be searched for. If embedding is set, search will use embedding instead of entity_id.
- NearestNeighborQuery.Embedding embedding = 2
  Optional. The embedding vector that be used for similar search.
int32 neighbor_count = 3
Optional. The number of similar entities to be retrieved from feature view for each query.
repeated NearestNeighborQuery.StringFilter string_filters = 4
Optional. The list of string filters.
repeated NearestNeighborQuery.NumericFilter numeric_filters = 8
Optional. The list of numeric filters.
int32 per_crowding_attribute_neighbor_count = 5
Optional. Crowding is a constraint on a neighbor list produced by nearest neighbor search requiring that no more than sper_crowding_attribute_neighbor_count of the k neighbors returned have the same value of crowding_attribute. It's used for improving result diversity.
optional NearestNeighborQuery.Parameters parameters = 7
Optional. Parameters that can be set to tune query on the fly.

The embedding vector.

Used in: NearestNeighborQuery

repeated float value = 1
Optional. Individual value in the embedding.

Numeric filter is used to search a subset of the entities by using boolean rules on numeric columns. For example: Database Point 0: {name: "a" value_int: 42} {name: "b" value_float: 1.0} Database Point 1: {name: "a" value_int: 10} {name: "b" value_float: 2.0} Database Point 2: {name: "a" value_int: -1} {name: "b" value_float: 3.0} Query: {name: "a" value_int: 12 operator: LESS} // Matches Point 1, 2 {name: "b" value_float: 2.0 operator: EQUAL} // Matches Point 1

Used in: NearestNeighborQuery

oneof Value
The type of Value must be consistent for all datapoints with a given name. This is verified at runtime.
- int64 value_int = 2
  int value type.
- float value_float = 3
  float value type.
- double value_double = 4
  double value type.
string name = 1
Required. Column name in BigQuery that used as filters.
optional NumericFilter.Operator op = 5
Optional. This MUST be specified for queries and must NOT be specified for database points.

Datapoints for which Operator is true relative to the query's Value field will be allowlisted.

Used in: NumericFilter

OPERATOR_UNSPECIFIED = 0
Unspecified operator.
LESS = 1
Entities are eligible if their value is < the query's.
LESS_EQUAL = 2
Entities are eligible if their value is <= the query's.
EQUAL = 3
Entities are eligible if their value is == the query's.
GREATER_EQUAL = 4
Entities are eligible if their value is >= the query's.
GREATER = 5
Entities are eligible if their value is > the query's.
NOT_EQUAL = 6
Entities are eligible if their value is != the query's.

Parameters that can be overrided in each query to tune query latency and recall.

Used in: NearestNeighborQuery

int32 approximate_neighbor_candidates = 1
Optional. The number of neighbors to find via approximate search before exact reordering is performed; if set, this value must be > neighbor_count.
double leaf_nodes_search_fraction = 2
Optional. The fraction of the number of leaves to search, set at query time allows user to tune search performance. This value increase result in both search accuracy and latency increase. The value should be between 0.0 and 1.0.

String filter is used to search a subset of the entities by using boolean rules on string columns. For example: if a query specifies string filter with 'name = color, allow_tokens = {red, blue}, deny_tokens = {purple}',' then that query will match entities that are red or blue, but if those points are also purple, then they will be excluded even if they are red/blue. Only string filter is supported for now, numeric filter will be supported in the near future.

Used in: NearestNeighborQuery

string name = 1
Required. Column names in BigQuery that used as filters.
repeated string allow_tokens = 2
Optional. The allowed tokens.
repeated string deny_tokens = 3
Optional. The denied tokens.

Runtime operation metadata with regard to Matching Engine Index.

Used in: CreateIndexOperationMetadata, UpdateIndexOperationMetadata

repeated NearestNeighborSearchOperationMetadata.ContentValidationStats content_validation_stats = 1
The validation stats of the content (per file) to be inserted or updated on the Matching Engine Index resource. Populated if contentsDeltaUri is provided as part of [Index.metadata][google.cloud.aiplatform.v1.Index.metadata]. Please note that, currently for those files that are broken or has unsupported file format, we will not have the stats for those files.
int64 data_bytes_count = 2
The ingested data size in bytes.

Used in: NearestNeighborSearchOperationMetadata

string source_gcs_uri = 1
Cloud Storage URI pointing to the original file in user's bucket.
int64 valid_record_count = 2
Number of records in this file that were successfully processed.
int64 invalid_record_count = 3
Number of records in this file we skipped due to validate errors.
repeated RecordError partial_errors = 4
The detail information of the partial failures encountered for those invalid records that couldn't be parsed. Up to 50 partial errors will be reported.
int64 valid_sparse_record_count = 5
Number of sparse records in this file that were successfully processed.
int64 invalid_sparse_record_count = 6
Number of sparse records in this file we skipped due to validate errors.

Used in: ContentValidationStats

RecordError.RecordErrorType error_type = 1
The error type of this record.
string error_message = 2
A human-readable message that is shown to the user to help them fix the error. Note that this message may change from time to time, your code should check against error_type as the source of truth.
string source_gcs_uri = 3
Cloud Storage URI pointing to the original file in user's bucket.
string embedding_id = 4
Empty if the embedding id is failed to parse.
string raw_record = 5
The original content of this record.

Used in: RecordError

ERROR_TYPE_UNSPECIFIED = 0
Default, shall not be used.
EMPTY_LINE = 1
The record is empty.
INVALID_JSON_SYNTAX = 2
Invalid json format.
INVALID_CSV_SYNTAX = 3
Invalid csv format.
INVALID_AVRO_SYNTAX = 4
Invalid avro format.
INVALID_EMBEDDING_ID = 5
The embedding id is not valid.
EMBEDDING_SIZE_MISMATCH = 6
The size of the dense embedding vectors does not match with the specified dimension.
NAMESPACE_MISSING = 7
The `namespace` field is missing.
PARSING_ERROR = 8
Generic catch-all error. Only used for validation failure where the root cause cannot be easily retrieved programmatically.
DUPLICATE_NAMESPACE = 9
There are multiple restricts with the same `namespace` value.
OP_IN_DATAPOINT = 10
Numeric restrict has operator specified in datapoint.
MULTIPLE_VALUES = 11
Numeric restrict has multiple values specified.
INVALID_NUMERIC_VALUE = 12
Numeric restrict has invalid numeric value specified.
INVALID_ENCODING = 13
File is not in UTF_8 format.
INVALID_SPARSE_DIMENSIONS = 14
Error parsing sparse dimensions field.
INVALID_TOKEN_VALUE = 15
Token restrict value is invalid.
INVALID_SPARSE_EMBEDDING = 16
Invalid sparse embedding.
INVALID_EMBEDDING = 17
Invalid dense embedding.

Nearest neighbors for one query.

Used in: SearchNearestEntitiesResponse

repeated NearestNeighbors.Neighbor neighbors = 1
All its neighbors.

A neighbor of the query vector.

Used in: NearestNeighbors

string entity_id = 1
The id of the similar entity.
double distance = 2
The distance between the neighbor and the query vector.
optional FetchFeatureValuesResponse entity_key_values = 3
The attributes of the neighbor, e.g. filters, crowding and metadata Note that full entities are returned only when "return_full_entity" is set to true. Otherwise, only the "entity_id" and "distance" fields are populated.

Neighbors for example-based explanations.

Used in: Explanation

string neighbor_id = 1
Output only. The neighbor id.
double neighbor_distance = 2
Output only. The neighbor distance.

Network spec.

Used in: NotebookExecutionJob.CustomEnvironmentSpec, NotebookRuntime, NotebookRuntimeTemplate

bool enable_internet_access = 1
Whether to enable public internet access. Default false.
string network = 2
The full name of the Google Compute Engine [network](https://cloud.google.com//compute/docs/networks-and-firewalls#networks)
string subnetwork = 3
The name of the subnet that this instance is in. Format: `projects/{project_id_or_number}/regions/{region}/subnetworks/{subnetwork_id}`

Represents a mount configuration for Network File System (NFS) to mount.

Used in: WorkerPoolSpec

string server = 1
Required. IP address of the NFS server.
string path = 2
Required. Source path exported from NFS server. Has to start with '/', and combined with the ip address, it indicates the source mount path in the form of `server:path`
string mount_point = 3
Required. Destination mount path. The NFS will be mounted for the user under /mnt/nfs/<mount_point>

The euc configuration of NotebookRuntimeTemplate.

bool euc_disabled = 1
Input only. Whether EUC is disabled in this NotebookRuntimeTemplate. In proto3, the default value of a boolean is false. In this way, by default EUC will be enabled for NotebookRuntimeTemplate.
bool bypass_actas_check = 2
Output only. Whether ActAs check is bypassed for service account attached to the VM. If false, we need ActAs check for the default Compute Engine Service account. When a Runtime is created, a VM is allocated using Default Compute Engine Service Account. Any user requesting to use this Runtime requires Service Account User (ActAs) permission over this SA. If true, Runtime owner is using EUC and does not require the above permission as VM no longer use default Compute Engine SA, but a P4SA.

NotebookExecutionJob represents an instance of a notebook execution.

Used as response type in: NotebookService.GetNotebookExecutionJob

Used as field type in: CreateNotebookExecutionJobRequest, ListNotebookExecutionJobsResponse

oneof notebook_source
The input notebook.
- NotebookExecutionJob.DataformRepositorySource dataform_repository_source = 3
  The Dataform Repository pointing to a single file notebook repository.
- NotebookExecutionJob.GcsNotebookSource gcs_notebook_source = 4
  The Cloud Storage url pointing to the ipynb file. Format: `gs://bucket/notebook_file.ipynb`
- NotebookExecutionJob.DirectNotebookSource direct_notebook_source = 17
  The contents of an input notebook file.
oneof environment_spec
The compute config to use for an execution job.
- string notebook_runtime_template_resource_name = 14
  The NotebookRuntimeTemplate to source compute configuration from.
- NotebookExecutionJob.CustomEnvironmentSpec custom_environment_spec = 16
  The custom compute configuration for an execution job.
oneof execution_sink
The location to store the notebook execution result.
- string gcs_output_uri = 8
  The Cloud Storage location to upload the result to. Format: `gs://bucket-name`
oneof execution_identity
The identity to run the execution as.
- string execution_user = 9
  The user email to run the execution as. Only supported by Colab runtimes.
- string service_account = 18
  The service account to run the execution as.
oneof runtime_environment
Runtime environment for the notebook execution job. If unspecified, the default runtime of Colab is used.
- NotebookExecutionJob.WorkbenchRuntime workbench_runtime = 23
  The Workbench runtime configuration to use for the notebook execution.
string name = 1
Output only. The resource name of this NotebookExecutionJob. Format: `projects/{project_id}/locations/{location}/notebookExecutionJobs/{job_id}`
string display_name = 2
The display name of the NotebookExecutionJob. The name can be up to 128 characters long and can consist of any UTF-8 characters.
optional protobuf.Duration execution_timeout = 5
Max running time of the execution job in seconds (default 86400s / 24 hrs).
string schedule_resource_name = 6
Output only. The Schedule resource name if this job is triggered by one. Format: `projects/{project_id}/locations/{location}/schedules/{schedule_id}`
JobState job_state = 10
Output only. The state of the NotebookExecutionJob.
optional rpc.Status status = 11
Output only. Populated when the NotebookExecutionJob is completed. When there is an error during notebook execution, the error details are populated.
optional protobuf.Timestamp create_time = 12
Output only. Timestamp when this NotebookExecutionJob was created.
optional protobuf.Timestamp update_time = 13
Output only. Timestamp when this NotebookExecutionJob was most recently updated.
map<string, string> labels = 19
The labels with user-defined metadata to organize NotebookExecutionJobs. 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. System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
string kernel_name = 20
The name of the kernel to use during notebook execution. If unset, the default kernel is used.
optional EncryptionSpec encryption_spec = 22
Customer-managed encryption key spec for the notebook execution job. This field is auto-populated if the [NotebookRuntimeTemplate][google.cloud.aiplatform.v1.NotebookRuntimeTemplate] has an encryption spec.

Compute configuration to use for an execution job.

Used in: NotebookExecutionJob

optional MachineSpec machine_spec = 1
The specification of a single machine for the execution job.
optional PersistentDiskSpec persistent_disk_spec = 2
The specification of a persistent disk to attach for the execution job.
optional NetworkSpec network_spec = 3
The network configuration to use for the execution job.

The Dataform Repository containing the input notebook.

Used in: NotebookExecutionJob

string dataform_repository_resource_name = 1
The resource name of the Dataform Repository. Format: `projects/{project_id}/locations/{location}/repositories/{repository_id}`
string commit_sha = 2
The commit SHA to read repository with. If unset, the file will be read at HEAD.

The content of the input notebook in ipynb format.

Used in: NotebookExecutionJob

bytes content = 1
The base64-encoded contents of the input notebook file.

The Cloud Storage uri for the input notebook.

Used in: NotebookExecutionJob

string uri = 1
The Cloud Storage uri pointing to the ipynb file. Format: `gs://bucket/notebook_file.ipynb`
string generation = 2
The version of the Cloud Storage object to read. If unset, the current version of the object is read. See https://cloud.google.com/storage/docs/metadata#generation-number.

Configuration for a Workbench Instances-based environment.

Used in: NotebookExecutionJob

(message has no fields)

Views for Get/List NotebookExecutionJob

Used in: GetNotebookExecutionJobRequest, ListNotebookExecutionJobsRequest

NOTEBOOK_EXECUTION_JOB_VIEW_UNSPECIFIED = 0
When unspecified, the API defaults to the BASIC view.
NOTEBOOK_EXECUTION_JOB_VIEW_BASIC = 1
Includes all fields except for direct notebook inputs.
NOTEBOOK_EXECUTION_JOB_VIEW_FULL = 2
Includes all fields.

The idle shutdown configuration of NotebookRuntimeTemplate, which contains the idle_timeout as required field.

optional protobuf.Duration idle_timeout = 1
Required. Duration is accurate to the second. In Notebook, Idle Timeout is accurate to minute so the range of idle_timeout (second) is: 10 * 60 ~ 1440 * 60.
bool idle_shutdown_disabled = 2
Whether Idle Shutdown is disabled in this NotebookRuntimeTemplate.

A runtime is a virtual machine allocated to a particular user for a particular Notebook file on temporary basis with lifetime limited to 24 hours.

Used as response type in: NotebookService.GetNotebookRuntime

Used as field type in: AssignNotebookRuntimeRequest, ListNotebookRuntimesResponse

string name = 1
Output only. The resource name of the NotebookRuntime.
string runtime_user = 2
Required. The user email of the NotebookRuntime.
optional NotebookRuntimeTemplateRef notebook_runtime_template_ref = 3
Output only. The pointer to NotebookRuntimeTemplate this NotebookRuntime is created from.
string proxy_uri = 5
Output only. The proxy endpoint used to access the NotebookRuntime.
optional protobuf.Timestamp create_time = 6
Output only. Timestamp when this NotebookRuntime was created.
optional protobuf.Timestamp update_time = 7
Output only. Timestamp when this NotebookRuntime was most recently updated.
NotebookRuntime.HealthState health_state = 8
Output only. The health state of the NotebookRuntime.
string display_name = 10
Required. The display name of the NotebookRuntime. The name can be up to 128 characters long and can consist of any UTF-8 characters.
string description = 11
The description of the NotebookRuntime.
string service_account = 13
Output only. Deprecated: This field is no longer used and the "Vertex AI Notebook Service Account" (service-PROJECT_NUMBER@gcp-sa-aiplatform-vm.iam.gserviceaccount.com) is used for the runtime workload identity. See https://cloud.google.com/iam/docs/service-agents#vertex-ai-notebook-service-account for more details. The service account that the NotebookRuntime workload runs as.
NotebookRuntime.RuntimeState runtime_state = 14
Output only. The runtime (instance) state of the NotebookRuntime.
bool is_upgradable = 15
Output only. Whether NotebookRuntime is upgradable.
map<string, string> labels = 16
The labels with user-defined metadata to organize your NotebookRuntime. 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. No more than 64 user labels can be associated with one NotebookRuntime (System labels are excluded). See https://goo.gl/xmQnxf for more information and examples of labels. System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable. Following system labels exist for NotebookRuntime: * "aiplatform.googleapis.com/notebook_runtime_gce_instance_id": output only, its value is the Compute Engine instance id. * "aiplatform.googleapis.com/colab_enterprise_entry_service": its value is either "bigquery" or "vertex"; if absent, it should be "vertex". This is to describe the entry service, either BigQuery or Vertex.
optional protobuf.Timestamp expiration_time = 17
Output only. Timestamp when this NotebookRuntime will be expired: 1. System Predefined NotebookRuntime: 24 hours after creation. After expiration, system predifined runtime will be deleted. 2. User created NotebookRuntime: 6 months after last upgrade. After expiration, user created runtime will be stopped and allowed for upgrade.
string version = 18
Output only. The VM os image version of NotebookRuntime.
NotebookRuntimeType notebook_runtime_type = 19
Output only. The type of the notebook runtime.
optional MachineSpec machine_spec = 20
Output only. The specification of a single machine used by the notebook runtime.
optional PersistentDiskSpec data_persistent_disk_spec = 21
Output only. The specification of [persistent disk][https://cloud.google.com/compute/docs/disks/persistent-disks] attached to the notebook runtime as data disk storage.
optional NetworkSpec network_spec = 22
Output only. Network spec of the notebook runtime.
optional NotebookIdleShutdownConfig idle_shutdown_config = 23
Output only. The idle shutdown configuration of the notebook runtime.
optional NotebookEucConfig euc_config = 24
Output only. EUC configuration of the notebook runtime.
optional ShieldedVmConfig shielded_vm_config = 32
Output only. Runtime Shielded VM spec.
repeated string network_tags = 25
Optional. The Compute Engine tags to add to runtime (see [Tagging instances](https://cloud.google.com/vpc/docs/add-remove-network-tags)).
optional NotebookSoftwareConfig software_config = 31
Output only. Software config of the notebook runtime.
optional EncryptionSpec encryption_spec = 28
Output only. Customer-managed encryption key spec for the notebook runtime.
bool satisfies_pzs = 29
Output only. Reserved for future use.
bool satisfies_pzi = 30
Output only. Reserved for future use.

The substate of the NotebookRuntime to display health information.

Used in: NotebookRuntime

HEALTH_STATE_UNSPECIFIED = 0
Unspecified health state.
HEALTHY = 1
NotebookRuntime is in healthy state. Applies to ACTIVE state.
UNHEALTHY = 2
NotebookRuntime is in unhealthy state. Applies to ACTIVE state.

The substate of the NotebookRuntime to display state of runtime. The resource of NotebookRuntime is in ACTIVE state for these sub state.

Used in: NotebookRuntime

RUNTIME_STATE_UNSPECIFIED = 0
Unspecified runtime state.
RUNNING = 1
NotebookRuntime is in running state.
BEING_STARTED = 2
NotebookRuntime is in starting state.
BEING_STOPPED = 3
NotebookRuntime is in stopping state.
STOPPED = 4
NotebookRuntime is in stopped state.
BEING_UPGRADED = 5
NotebookRuntime is in upgrading state. It is in the middle of upgrading process.
ERROR = 100
NotebookRuntime was unable to start/stop properly.
INVALID = 101
NotebookRuntime is in invalid state. Cannot be recovered.

A template that specifies runtime configurations such as machine type, runtime version, network configurations, etc. Multiple runtimes can be created from a runtime template.

Used as response type in: NotebookService.GetNotebookRuntimeTemplate, NotebookService.UpdateNotebookRuntimeTemplate

Used as field type in: CreateNotebookRuntimeTemplateRequest, ListNotebookRuntimeTemplatesResponse, UpdateNotebookRuntimeTemplateRequest

string name = 1
The resource name of the NotebookRuntimeTemplate.
string display_name = 2
Required. The display name of the NotebookRuntimeTemplate. The name can be up to 128 characters long and can consist of any UTF-8 characters.
string description = 3
The description of the NotebookRuntimeTemplate.
bool is_default = 4
Output only. Deprecated: This field has no behavior. Use notebook_runtime_type = 'ONE_CLICK' instead. The default template to use if not specified.
optional MachineSpec machine_spec = 5
Optional. Immutable. The specification of a single machine for the template.
optional PersistentDiskSpec data_persistent_disk_spec = 8
Optional. The specification of [persistent disk][https://cloud.google.com/compute/docs/disks/persistent-disks] attached to the runtime as data disk storage.
optional NetworkSpec network_spec = 12
Optional. Network spec.
string service_account = 13
Deprecated: This field is ignored and the "Vertex AI Notebook Service Account" (service-PROJECT_NUMBER@gcp-sa-aiplatform-vm.iam.gserviceaccount.com) is used for the runtime workload identity. See https://cloud.google.com/iam/docs/service-agents#vertex-ai-notebook-service-account for more details. For NotebookExecutionJob, use NotebookExecutionJob.service_account instead. The service account that the runtime workload runs as. You can use any service account within the same project, but you must have the service account user permission to use the instance. If not specified, the [Compute Engine default service account](https://cloud.google.com/compute/docs/access/service-accounts#default_service_account) is used.
string etag = 14
Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
map<string, string> labels = 15
The labels with user-defined metadata to organize the NotebookRuntimeTemplates. 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.
optional NotebookIdleShutdownConfig idle_shutdown_config = 17
The idle shutdown configuration of NotebookRuntimeTemplate. This config will only be set when idle shutdown is enabled.
optional NotebookEucConfig euc_config = 18
EUC configuration of the NotebookRuntimeTemplate.
optional protobuf.Timestamp create_time = 10
Output only. Timestamp when this NotebookRuntimeTemplate was created.
optional protobuf.Timestamp update_time = 11
Output only. Timestamp when this NotebookRuntimeTemplate was most recently updated.
NotebookRuntimeType notebook_runtime_type = 19
Optional. Immutable. The type of the notebook runtime template.
optional ShieldedVmConfig shielded_vm_config = 20
Optional. Immutable. Runtime Shielded VM spec.
repeated string network_tags = 21
Optional. The Compute Engine tags to add to runtime (see [Tagging instances](https://cloud.google.com/vpc/docs/add-remove-network-tags)).
optional EncryptionSpec encryption_spec = 23
Customer-managed encryption key spec for the notebook runtime.
optional NotebookSoftwareConfig software_config = 24
Optional. The notebook software configuration of the notebook runtime.

Points to a NotebookRuntimeTemplateRef.

Used in: NotebookRuntime

string notebook_runtime_template = 1
Immutable. A resource name of the NotebookRuntimeTemplate.

Represents a notebook runtime type.

NOTEBOOK_RUNTIME_TYPE_UNSPECIFIED = 0
Unspecified notebook runtime type, NotebookRuntimeType will default to USER_DEFINED.
USER_DEFINED = 1
runtime or template with coustomized configurations from user.
ONE_CLICK = 2
runtime or template with system defined configurations.

Notebook Software Config.

repeated EnvVar env = 1
Optional. Environment variables to be passed to the container. Maximum limit is 100.
optional PostStartupScriptConfig post_startup_script_config = 2
Optional. Post startup script config.

PSC config that is used to automatically create forwarding rule via ServiceConnectionMap.

Used in: DeployedIndex

string project_id = 1
Required. Project id used to create forwarding rule.
string network = 2
Required. The full name of the Google Compute Engine [network](https://cloud.google.com/compute/docs/networks-and-firewalls#networks). [Format](https://cloud.google.com/compute/docs/reference/rest/v1/networks/insert): `projects/{project}/global/networks/{network}`. Where {project} is a project number, as in '12345', and {network} is network name.

Pairwise prediction autorater preference.

Used in: PairwiseMetricResult, PairwiseQuestionAnsweringQualityResult, PairwiseSummarizationQualityResult

PAIRWISE_CHOICE_UNSPECIFIED = 0
Unspecified prediction choice.
BASELINE = 1
Baseline prediction wins
CANDIDATE = 2
Candidate prediction wins
TIE = 3
Winner cannot be determined

Input for pairwise metric.

optional PairwiseMetricSpec metric_spec = 1
Required. Spec for pairwise metric.
optional PairwiseMetricInstance instance = 2
Required. Pairwise metric instance.

Pairwise metric instance. Usually one instance corresponds to one row in an evaluation dataset.

Used in: PairwiseMetricInput

oneof instance
Instance for pairwise metric.
- string json_instance = 1
  Instance specified as a json string. String key-value pairs are expected in the json_instance to render PairwiseMetricSpec.instance_prompt_template.

Spec for pairwise metric result.

PairwiseChoice pairwise_choice = 1
Output only. Pairwise metric choice.
string explanation = 2
Output only. Explanation for pairwise metric score.

Spec for pairwise metric.

Used in: PairwiseMetricInput

optional string metric_prompt_template = 1
Required. Metric prompt template for pairwise metric.

Input for pairwise question answering quality metric.

optional PairwiseQuestionAnsweringQualitySpec metric_spec = 1
Required. Spec for pairwise question answering quality score metric.
optional PairwiseQuestionAnsweringQualityInstance instance = 2
Required. Pairwise question answering quality instance.

Spec for pairwise question answering quality instance.

Used in: PairwiseQuestionAnsweringQualityInput

optional string prediction = 1
Required. Output of the candidate model.
optional string baseline_prediction = 2
Required. Output of the baseline model.
optional string reference = 3
Optional. Ground truth used to compare against the prediction.
optional string context = 4
Required. Text to answer the question.
optional string instruction = 5
Required. Question Answering prompt for LLM.

Spec for pairwise question answering quality result.

PairwiseChoice pairwise_choice = 1
Output only. Pairwise question answering prediction choice.
string explanation = 2
Output only. Explanation for question answering quality score.
optional float confidence = 3
Output only. Confidence for question answering quality score.

Spec for pairwise question answering quality score metric.

Used in: PairwiseQuestionAnsweringQualityInput

bool use_reference = 1
Optional. Whether to use instance.reference to compute question answering quality.
int32 version = 2
Optional. Which version to use for evaluation.

Input for pairwise summarization quality metric.

optional PairwiseSummarizationQualitySpec metric_spec = 1
Required. Spec for pairwise summarization quality score metric.
optional PairwiseSummarizationQualityInstance instance = 2
Required. Pairwise summarization quality instance.

Spec for pairwise summarization quality instance.

Used in: PairwiseSummarizationQualityInput

optional string prediction = 1
Required. Output of the candidate model.
optional string baseline_prediction = 2
Required. Output of the baseline model.
optional string reference = 3
Optional. Ground truth used to compare against the prediction.
optional string context = 4
Required. Text to be summarized.
optional string instruction = 5
Required. Summarization prompt for LLM.

Spec for pairwise summarization quality result.

PairwiseChoice pairwise_choice = 1
Output only. Pairwise summarization prediction choice.
string explanation = 2
Output only. Explanation for summarization quality score.
optional float confidence = 3
Output only. Confidence for summarization quality score.

Spec for pairwise summarization quality score metric.

Used in: PairwiseSummarizationQualityInput

bool use_reference = 1
Optional. Whether to use instance.reference to compute pairwise summarization quality.
int32 version = 2
Optional. Which version to use for evaluation.

A datatype containing media that is part of a multi-part `Content` message. A `Part` consists of data which has an associated datatype. A `Part` can only contain one of the accepted types in `Part.data`. A `Part` must have a fixed IANA MIME type identifying the type and subtype of the media if `inline_data` or `file_data` field is filled with raw bytes.

Used in: Content

oneof data
- string text = 1
  Optional. Text part (can be code).
- Blob inline_data = 2
  Optional. Inlined bytes data.
- FileData file_data = 3
  Optional. URI based data.
- FunctionCall function_call = 5
  Optional. A predicted [FunctionCall] returned from the model that contains a string representing the [FunctionDeclaration.name] with the parameters and their values.
- FunctionResponse function_response = 6
  Optional. The result output of a [FunctionCall] that contains a string representing the [FunctionDeclaration.name] and a structured JSON object containing any output from the function call. It is used as context to the model.
- ExecutableCode executable_code = 8
  Optional. Code generated by the model that is meant to be executed.
- CodeExecutionResult code_execution_result = 9
  Optional. Result of executing the [ExecutableCode].
oneof metadata
- VideoMetadata video_metadata = 4
  Optional. Video metadata. The metadata should only be specified while the video data is presented in inline_data or file_data.

Represents the spec of [persistent disk][https://cloud.google.com/compute/docs/disks/persistent-disks] options.

Used in: NotebookExecutionJob.CustomEnvironmentSpec, NotebookRuntime, NotebookRuntimeTemplate

string disk_type = 1
Type of the disk (default is "pd-standard"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) "pd-standard" (Persistent Disk Hard Disk Drive) "pd-balanced" (Balanced Persistent Disk) "pd-extreme" (Extreme Persistent Disk)
int64 disk_size_gb = 2
Size in GB of the disk (default is 100GB).

Represents long-lasting resources that are dedicated to users to runs custom workloads. A PersistentResource can have multiple node pools and each node pool can have its own machine spec.

Used as response type in: PersistentResourceService.GetPersistentResource

Used as field type in: CreatePersistentResourceRequest, ListPersistentResourcesResponse, UpdatePersistentResourceRequest

string name = 1
Immutable. Resource name of a PersistentResource.
string display_name = 2
Optional. The display name of the PersistentResource. The name can be up to 128 characters long and can consist of any UTF-8 characters.
repeated ResourcePool resource_pools = 4
Required. The spec of the pools of different resources.
PersistentResource.State state = 5
Output only. The detailed state of a Study.
optional rpc.Status error = 6
Output only. Only populated when persistent resource's state is `STOPPING` or `ERROR`.
optional protobuf.Timestamp create_time = 7
Output only. Time when the PersistentResource was created.
optional protobuf.Timestamp start_time = 8
Output only. Time when the PersistentResource for the first time entered the `RUNNING` state.
optional protobuf.Timestamp update_time = 9
Output only. Time when the PersistentResource was most recently updated.
map<string, string> labels = 10
Optional. The labels with user-defined metadata to organize PersistentResource. 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.
string network = 11
Optional. The full name of the Compute Engine [network](/compute/docs/networks-and-firewalls#networks) to peered with Vertex AI to host the persistent resources. For example, `projects/12345/global/networks/myVPC`. [Format](/compute/docs/reference/rest/v1/networks/insert) is of the form `projects/{project}/global/networks/{network}`. Where {project} is a project number, as in `12345`, and {network} is a network name. To specify this field, you must have already [configured VPC Network Peering for Vertex AI](https://cloud.google.com/vertex-ai/docs/general/vpc-peering). If this field is left unspecified, the resources aren't peered with any network.
optional EncryptionSpec encryption_spec = 12
Optional. Customer-managed encryption key spec for a PersistentResource. If set, this PersistentResource and all sub-resources of this PersistentResource will be secured by this key.
optional ResourceRuntimeSpec resource_runtime_spec = 13
Optional. Persistent Resource runtime spec. For example, used for Ray cluster configuration.
optional ResourceRuntime resource_runtime = 14
Output only. Runtime information of the Persistent Resource.
repeated string reserved_ip_ranges = 15
Optional. A list of names for the reserved IP ranges under the VPC network that can be used for this persistent resource. If set, we will deploy the persistent resource within the provided IP ranges. Otherwise, the persistent resource is deployed to any IP ranges under the provided VPC network. Example: ['vertex-ai-ip-range'].

Describes the PersistentResource state.

Used in: PersistentResource

STATE_UNSPECIFIED = 0
Not set.
PROVISIONING = 1
The PROVISIONING state indicates the persistent resources is being created.
RUNNING = 3
The RUNNING state indicates the persistent resource is healthy and fully usable.
STOPPING = 4
The STOPPING state indicates the persistent resource is being deleted.
ERROR = 5
The ERROR state indicates the persistent resource may be unusable. Details can be found in the `error` field.
REBOOTING = 6
The REBOOTING state indicates the persistent resource is being rebooted (PR is not available right now but is expected to be ready again later).
UPDATING = 7
The UPDATING state indicates the persistent resource is being updated.

Represents the failure policy of a pipeline. Currently, the default of a pipeline is that the pipeline will continue to run until no more tasks can be executed, also known as PIPELINE_FAILURE_POLICY_FAIL_SLOW. However, if a pipeline is set to PIPELINE_FAILURE_POLICY_FAIL_FAST, it will stop scheduling any new tasks when a task has failed. Any scheduled tasks will continue to completion.

Used in: PipelineJob.RuntimeConfig

PIPELINE_FAILURE_POLICY_UNSPECIFIED = 0
Default value, and follows fail slow behavior.
PIPELINE_FAILURE_POLICY_FAIL_SLOW = 1
Indicates that the pipeline should continue to run until all possible tasks have been scheduled and completed.
PIPELINE_FAILURE_POLICY_FAIL_FAST = 2
Indicates that the pipeline should stop scheduling new tasks after a task has failed.

An instance of a machine learning PipelineJob.

Used as response type in: PipelineService.CreatePipelineJob, PipelineService.GetPipelineJob

Used as field type in: BatchCancelPipelineJobsResponse, BatchDeletePipelineJobsResponse, CreatePipelineJobRequest, ListPipelineJobsResponse

string name = 1
Output only. The resource name of the PipelineJob.
string display_name = 2
The display name of the Pipeline. The name can be up to 128 characters long and can consist of any UTF-8 characters.
optional protobuf.Timestamp create_time = 3
Output only. Pipeline creation time.
optional protobuf.Timestamp start_time = 4
Output only. Pipeline start time.
optional protobuf.Timestamp end_time = 5
Output only. Pipeline end time.
optional protobuf.Timestamp update_time = 6
Output only. Timestamp when this PipelineJob was most recently updated.
optional protobuf.Struct pipeline_spec = 7
The spec of the pipeline.
PipelineState state = 8
Output only. The detailed state of the job.
optional PipelineJobDetail job_detail = 9
Output only. The details of pipeline run. Not available in the list view.
optional rpc.Status error = 10
Output only. The error that occurred during pipeline execution. Only populated when the pipeline's state is FAILED or CANCELLED.
map<string, string> labels = 11
The labels with user-defined metadata to organize PipelineJob. 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. Note there is some reserved label key for Vertex AI Pipelines. - `vertex-ai-pipelines-run-billing-id`, user set value will get overrided.
optional PipelineJob.RuntimeConfig runtime_config = 12
Runtime config of the pipeline.
optional EncryptionSpec encryption_spec = 16
Customer-managed encryption key spec for a pipelineJob. If set, this PipelineJob and all of its sub-resources will be secured by this key.
string service_account = 17
The service account that the pipeline workload runs as. If not specified, the Compute Engine default service account in the project will be used. See https://cloud.google.com/compute/docs/access/service-accounts#default_service_account Users starting the pipeline must have the `iam.serviceAccounts.actAs` permission on this service account.
string network = 18
The full name of the Compute Engine [network](/compute/docs/networks-and-firewalls#networks) to which the Pipeline Job's workload should be peered. For example, `projects/12345/global/networks/myVPC`. [Format](/compute/docs/reference/rest/v1/networks/insert) is of the form `projects/{project}/global/networks/{network}`. Where {project} is a project number, as in `12345`, and {network} is a network name. Private services access must already be configured for the network. Pipeline job will apply the network configuration to the Google Cloud resources being launched, if applied, such as Vertex AI Training or Dataflow job. If left unspecified, the workload is not peered with any network.
repeated string reserved_ip_ranges = 25
A list of names for the reserved ip ranges under the VPC network that can be used for this Pipeline Job's workload. If set, we will deploy the Pipeline Job's workload within the provided ip ranges. Otherwise, the job will be deployed to any ip ranges under the provided VPC network. Example: ['vertex-ai-ip-range'].
string template_uri = 19
A template uri from where the [PipelineJob.pipeline_spec][google.cloud.aiplatform.v1.PipelineJob.pipeline_spec], if empty, will be downloaded. Currently, only uri from Vertex Template Registry & Gallery is supported. Reference to https://cloud.google.com/vertex-ai/docs/pipelines/create-pipeline-template.
optional PipelineTemplateMetadata template_metadata = 20
Output only. Pipeline template metadata. Will fill up fields if [PipelineJob.template_uri][google.cloud.aiplatform.v1.PipelineJob.template_uri] is from supported template registry.
string schedule_name = 22
Output only. The schedule resource name. Only returned if the Pipeline is created by Schedule API.
bool preflight_validations = 26
Optional. Whether to do component level validations before job creation.

The runtime config of a PipelineJob.

Used in: PipelineJob

map<string, Value> parameters = 1
Deprecated. Use [RuntimeConfig.parameter_values][google.cloud.aiplatform.v1.PipelineJob.RuntimeConfig.parameter_values] instead. The runtime parameters of the PipelineJob. The parameters will be passed into [PipelineJob.pipeline_spec][google.cloud.aiplatform.v1.PipelineJob.pipeline_spec] to replace the placeholders at runtime. This field is used by pipelines built using `PipelineJob.pipeline_spec.schema_version` 2.0.0 or lower, such as pipelines built using Kubeflow Pipelines SDK 1.8 or lower.
string gcs_output_directory = 2
Required. A path in a Cloud Storage bucket, which will be treated as the root output directory of the pipeline. It is used by the system to generate the paths of output artifacts. The artifact paths are generated with a sub-path pattern `{job_id}/{task_id}/{output_key}` under the specified output directory. The service account specified in this pipeline must have the `storage.objects.get` and `storage.objects.create` permissions for this bucket.
map<string, protobuf.Value> parameter_values = 3
The runtime parameters of the PipelineJob. The parameters will be passed into [PipelineJob.pipeline_spec][google.cloud.aiplatform.v1.PipelineJob.pipeline_spec] to replace the placeholders at runtime. This field is used by pipelines built using `PipelineJob.pipeline_spec.schema_version` 2.1.0, such as pipelines built using Kubeflow Pipelines SDK 1.9 or higher and the v2 DSL.
PipelineFailurePolicy failure_policy = 4
Represents the failure policy of a pipeline. Currently, the default of a pipeline is that the pipeline will continue to run until no more tasks can be executed, also known as PIPELINE_FAILURE_POLICY_FAIL_SLOW. However, if a pipeline is set to PIPELINE_FAILURE_POLICY_FAIL_FAST, it will stop scheduling any new tasks when a task has failed. Any scheduled tasks will continue to completion.
map<string, RuntimeConfig.InputArtifact> input_artifacts = 5
The runtime artifacts of the PipelineJob. The key will be the input artifact name and the value would be one of the InputArtifact.

The type of an input artifact.

Used in: RuntimeConfig

oneof kind
- string artifact_id = 1
  Artifact resource id from MLMD. Which is the last portion of an artifact resource name: `projects/{project}/locations/{location}/metadataStores/default/artifacts/{artifact_id}`. The artifact must stay within the same project, location and default metadatastore as the pipeline.

The runtime detail of PipelineJob.

Used in: PipelineJob

optional Context pipeline_context = 1
Output only. The context of the pipeline.
optional Context pipeline_run_context = 2
Output only. The context of the current pipeline run.
repeated PipelineTaskDetail task_details = 3
Output only. The runtime details of the tasks under the pipeline.

Describes the state of a pipeline.

Used in: PipelineJob, TrainingPipeline

PIPELINE_STATE_UNSPECIFIED = 0
The pipeline state is unspecified.
PIPELINE_STATE_QUEUED = 1
The pipeline has been created or resumed, and processing has not yet begun.
PIPELINE_STATE_PENDING = 2
The service is preparing to run the pipeline.
PIPELINE_STATE_RUNNING = 3
The pipeline is in progress.
PIPELINE_STATE_SUCCEEDED = 4
The pipeline completed successfully.
PIPELINE_STATE_FAILED = 5
The pipeline failed.
PIPELINE_STATE_CANCELLING = 6
The pipeline is being cancelled. From this state, the pipeline may only go to either PIPELINE_STATE_SUCCEEDED, PIPELINE_STATE_FAILED or PIPELINE_STATE_CANCELLED.
PIPELINE_STATE_CANCELLED = 7
The pipeline has been cancelled.
PIPELINE_STATE_PAUSED = 8
The pipeline has been stopped, and can be resumed.

The runtime detail of a task execution.

Used in: PipelineJobDetail

int64 task_id = 1
Output only. The system generated ID of the task.
int64 parent_task_id = 12
Output only. The id of the parent task if the task is within a component scope. Empty if the task is at the root level.
string task_name = 2
Output only. The user specified name of the task that is defined in [pipeline_spec][google.cloud.aiplatform.v1.PipelineJob.pipeline_spec].
optional protobuf.Timestamp create_time = 3
Output only. Task create time.
optional protobuf.Timestamp start_time = 4
Output only. Task start time.
optional protobuf.Timestamp end_time = 5
Output only. Task end time.
optional PipelineTaskExecutorDetail executor_detail = 6
Output only. The detailed execution info.
PipelineTaskDetail.State state = 7
Output only. State of the task.
optional Execution execution = 8
Output only. The execution metadata of the task.
optional rpc.Status error = 9
Output only. The error that occurred during task execution. Only populated when the task's state is FAILED or CANCELLED.
repeated PipelineTaskDetail.PipelineTaskStatus pipeline_task_status = 13
Output only. A list of task status. This field keeps a record of task status evolving over time.
map<string, PipelineTaskDetail.ArtifactList> inputs = 10
Output only. The runtime input artifacts of the task.
map<string, PipelineTaskDetail.ArtifactList> outputs = 11
Output only. The runtime output artifacts of the task.

A list of artifact metadata.

Used in: PipelineTaskDetail

repeated Artifact artifacts = 1
Output only. A list of artifact metadata.

A single record of the task status.

Used in: PipelineTaskDetail

optional protobuf.Timestamp update_time = 1
Output only. Update time of this status.
State state = 2
Output only. The state of the task.
optional rpc.Status error = 3
Output only. The error that occurred during the state. May be set when the state is any of the non-final state (PENDING/RUNNING/CANCELLING) or FAILED state. If the state is FAILED, the error here is final and not going to be retried. If the state is a non-final state, the error indicates a system-error being retried.

Specifies state of TaskExecution

Used in: PipelineTaskDetail, PipelineTaskStatus

STATE_UNSPECIFIED = 0
Unspecified.
PENDING = 1
Specifies pending state for the task.
RUNNING = 2
Specifies task is being executed.
SUCCEEDED = 3
Specifies task completed successfully.
CANCEL_PENDING = 4
Specifies Task cancel is in pending state.
CANCELLING = 5
Specifies task is being cancelled.
CANCELLED = 6
Specifies task was cancelled.
FAILED = 7
Specifies task failed.
SKIPPED = 8
Specifies task was skipped due to cache hit.
NOT_TRIGGERED = 9
Specifies that the task was not triggered because the task's trigger policy is not satisfied. The trigger policy is specified in the `condition` field of [PipelineJob.pipeline_spec][google.cloud.aiplatform.v1.PipelineJob.pipeline_spec].

The runtime detail of a pipeline executor.

Used in: PipelineTaskDetail

oneof details
- PipelineTaskExecutorDetail.ContainerDetail container_detail = 1
  Output only. The detailed info for a container executor.
- PipelineTaskExecutorDetail.CustomJobDetail custom_job_detail = 2
  Output only. The detailed info for a custom job executor.

The detail of a container execution. It contains the job names of the lifecycle of a container execution.

Used in: PipelineTaskExecutorDetail

string main_job = 1
Output only. The name of the [CustomJob][google.cloud.aiplatform.v1.CustomJob] for the main container execution.
string pre_caching_check_job = 2
Output only. The name of the [CustomJob][google.cloud.aiplatform.v1.CustomJob] for the pre-caching-check container execution. This job will be available if the [PipelineJob.pipeline_spec][google.cloud.aiplatform.v1.PipelineJob.pipeline_spec] specifies the `pre_caching_check` hook in the lifecycle events.
repeated string failed_main_jobs = 3
Output only. The names of the previously failed [CustomJob][google.cloud.aiplatform.v1.CustomJob] for the main container executions. The list includes the all attempts in chronological order.
repeated string failed_pre_caching_check_jobs = 4
Output only. The names of the previously failed [CustomJob][google.cloud.aiplatform.v1.CustomJob] for the pre-caching-check container executions. This job will be available if the [PipelineJob.pipeline_spec][google.cloud.aiplatform.v1.PipelineJob.pipeline_spec] specifies the `pre_caching_check` hook in the lifecycle events. The list includes the all attempts in chronological order.

The detailed info for a custom job executor.

Used in: PipelineTaskExecutorDetail

string job = 1
Output only. The name of the [CustomJob][google.cloud.aiplatform.v1.CustomJob].
repeated string failed_jobs = 3
Output only. The names of the previously failed [CustomJob][google.cloud.aiplatform.v1.CustomJob]. The list includes the all attempts in chronological order.

Pipeline template metadata if [PipelineJob.template_uri][google.cloud.aiplatform.v1.PipelineJob.template_uri] is from supported template registry. Currently, the only supported registry is Artifact Registry.

Used in: PipelineJob

string version = 3
The version_name in artifact registry. Will always be presented in output if the [PipelineJob.template_uri][google.cloud.aiplatform.v1.PipelineJob.template_uri] is from supported template registry. Format is "sha256:abcdef123456...".

Input for pointwise metric.

optional PointwiseMetricSpec metric_spec = 1
Required. Spec for pointwise metric.
optional PointwiseMetricInstance instance = 2
Required. Pointwise metric instance.

Pointwise metric instance. Usually one instance corresponds to one row in an evaluation dataset.

Used in: PointwiseMetricInput

oneof instance
Instance for pointwise metric.
- string json_instance = 1
  Instance specified as a json string. String key-value pairs are expected in the json_instance to render PointwiseMetricSpec.instance_prompt_template.

Spec for pointwise metric result.

optional float score = 1
Output only. Pointwise metric score.
string explanation = 2
Output only. Explanation for pointwise metric score.

Spec for pointwise metric.

Used in: PointwiseMetricInput

optional string metric_prompt_template = 1
Required. Metric prompt template for pointwise metric.

Represents a network port in a container.

Used in: ModelContainerSpec

int32 container_port = 3
The number of the port to expose on the pod's IP address. Must be a valid port number, between 1 and 65535 inclusive.

Post startup script config.

Used in: NotebookSoftwareConfig

string post_startup_script = 1
Optional. Post startup script to run after runtime is started.
string post_startup_script_url = 2
Optional. Post startup script url to download. Example: https://bucket/script.sh
PostStartupScriptConfig.PostStartupScriptBehavior post_startup_script_behavior = 3
Optional. Post startup script behavior that defines download and execution behavior.

Represents a notebook runtime post startup script behavior.

Used in: PostStartupScriptConfig

POST_STARTUP_SCRIPT_BEHAVIOR_UNSPECIFIED = 0
Unspecified post startup script behavior.
RUN_ONCE = 1
Run post startup script after runtime is started.
RUN_EVERY_START = 2
Run post startup script after runtime is stopped.
DOWNLOAD_AND_RUN_EVERY_START = 3
Download and run post startup script every time runtime is started.

Assigns input data to training, validation, and test sets based on the value of a provided key. Supported only for tabular Datasets.

Used in: InputDataConfig

string key = 1
Required. The key is a name of one of the Dataset's data columns. The value of the key (either the label's value or value in the column) must be one of {`training`, `validation`, `test`}, and it defines to which set the given piece of data is assigned. If for a piece of data the key is not present or has an invalid value, that piece is ignored by the pipeline.

Configuration for logging request-response to a BigQuery table.

Used in: Endpoint

bool enabled = 1
If logging is enabled or not.
double sampling_rate = 2
Percentage of requests to be logged, expressed as a fraction in range(0,1].
optional BigQueryDestination bigquery_destination = 3
BigQuery table for logging. If only given a project, a new dataset will be created with name `logging_<endpoint-display-name>_<endpoint-id>` where <endpoint-display-name> will be made BigQuery-dataset-name compatible (e.g. most special characters will become underscores). If no table name is given, a new table will be created with name `request_response_logging`

Contains the schemata used in Model's predictions and explanations via [PredictionService.Predict][google.cloud.aiplatform.v1.PredictionService.Predict], [PredictionService.Explain][google.cloud.aiplatform.v1.PredictionService.Explain] and [BatchPredictionJob][google.cloud.aiplatform.v1.BatchPredictionJob].

Used in: Model, PublisherModel, UnmanagedContainerModel

string instance_schema_uri = 1
Immutable. Points to a YAML file stored on Google Cloud Storage describing the format of a single instance, which are used in [PredictRequest.instances][google.cloud.aiplatform.v1.PredictRequest.instances], [ExplainRequest.instances][google.cloud.aiplatform.v1.ExplainRequest.instances] and [BatchPredictionJob.input_config][google.cloud.aiplatform.v1.BatchPredictionJob.input_config]. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). AutoML Models always have this field populated by Vertex AI. Note: The URI given on output will be immutable and probably different, including the URI scheme, than the one given on input. The output URI will point to a location where the user only has a read access.
string parameters_schema_uri = 2
Immutable. Points to a YAML file stored on Google Cloud Storage describing the parameters of prediction and explanation via [PredictRequest.parameters][google.cloud.aiplatform.v1.PredictRequest.parameters], [ExplainRequest.parameters][google.cloud.aiplatform.v1.ExplainRequest.parameters] and [BatchPredictionJob.model_parameters][google.cloud.aiplatform.v1.BatchPredictionJob.model_parameters]. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). AutoML Models always have this field populated by Vertex AI, if no parameters are supported, then it is set to an empty string. Note: The URI given on output will be immutable and probably different, including the URI scheme, than the one given on input. The output URI will point to a location where the user only has a read access.
string prediction_schema_uri = 3
Immutable. Points to a YAML file stored on Google Cloud Storage describing the format of a single prediction produced by this Model, which are returned via [PredictResponse.predictions][google.cloud.aiplatform.v1.PredictResponse.predictions], [ExplainResponse.explanations][google.cloud.aiplatform.v1.ExplainResponse.explanations], and [BatchPredictionJob.output_config][google.cloud.aiplatform.v1.BatchPredictionJob.output_config]. The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject). AutoML Models always have this field populated by Vertex AI. Note: The URI given on output will be immutable and probably different, including the URI scheme, than the one given on input. The output URI will point to a location where the user only has a read access.

Preset configuration for example-based explanations

Used in: Examples

optional Presets.Query query = 1
Preset option controlling parameters for speed-precision trade-off when querying for examples. If omitted, defaults to `PRECISE`.
Presets.Modality modality = 2
The modality of the uploaded model, which automatically configures the distance measurement and feature normalization for the underlying example index and queries. If your model does not precisely fit one of these types, it is okay to choose the closest type.

Preset option controlling parameters for different modalities

Used in: Presets

MODALITY_UNSPECIFIED = 0
Should not be set. Added as a recommended best practice for enums
IMAGE = 1
IMAGE modality
TEXT = 2
TEXT modality
TABULAR = 3
TABULAR modality

Preset option controlling parameters for query speed-precision trade-off

Used in: Presets

PRECISE = 0
More precise neighbors as a trade-off against slower response.
FAST = 1
Faster response as a trade-off against less precise neighbors.

PrivateEndpoints proto is used to provide paths for users to send requests privately. To send request via private service access, use predict_http_uri, explain_http_uri or health_http_uri. To send request via private service connect, use service_attachment.

Used in: DeployedModel

string predict_http_uri = 1
Output only. Http(s) path to send prediction requests.
string explain_http_uri = 2
Output only. Http(s) path to send explain requests.
string health_http_uri = 3
Output only. Http(s) path to send health check requests.
string service_attachment = 4
Output only. The name of the service attachment resource. Populated if private service connect is enabled.

Represents configuration for private service connect.

Used in: Endpoint, FeatureOnlineStore.DedicatedServingEndpoint, IndexEndpoint

bool enable_private_service_connect = 1
Required. If true, expose the IndexEndpoint via private service connect.
repeated string project_allowlist = 2
A list of Projects from which the forwarding rule will target the service attachment.
string service_attachment = 5
Output only. The name of the generated service attachment resource. This is only populated if the endpoint is deployed with PrivateServiceConnect.

Probe describes a health check to be performed against a container to determine whether it is alive or ready to receive traffic.

Used in: ModelContainerSpec

oneof probe_type
- Probe.ExecAction exec = 1
  ExecAction probes the health of a container by executing a command.
- Probe.HttpGetAction http_get = 4
  HttpGetAction probes the health of a container by sending an HTTP GET request.
- Probe.GrpcAction grpc = 5
  GrpcAction probes the health of a container by sending a gRPC request.
- Probe.TcpSocketAction tcp_socket = 6
  TcpSocketAction probes the health of a container by opening a TCP socket connection.
int32 period_seconds = 2
How often (in seconds) to perform the probe. Default to 10 seconds. Minimum value is 1. Must be less than timeout_seconds. Maps to Kubernetes probe argument 'periodSeconds'.
int32 timeout_seconds = 3
Number of seconds after which the probe times out. Defaults to 1 second. Minimum value is 1. Must be greater or equal to period_seconds. Maps to Kubernetes probe argument 'timeoutSeconds'.
int32 failure_threshold = 7
Number of consecutive failures before the probe is considered failed. Defaults to 3. Minimum value is 1. Maps to Kubernetes probe argument 'failureThreshold'.
int32 success_threshold = 8
Number of consecutive successes before the probe is considered successful. Defaults to 1. Minimum value is 1. Maps to Kubernetes probe argument 'successThreshold'.
int32 initial_delay_seconds = 9
Number of seconds to wait before starting the probe. Defaults to 0. Minimum value is 0. Maps to Kubernetes probe argument 'initialDelaySeconds'.

ExecAction specifies a command to execute.

Used in: Probe

repeated string command = 1
Command is the command line to execute inside the container, the working directory for the command is root ('/') in the container's filesystem. The command is simply exec'd, it is not run inside a shell, so traditional shell instructions ('|', etc) won't work. To use a shell, you need to explicitly call out to that shell. Exit status of 0 is treated as live/healthy and non-zero is unhealthy.

GrpcAction checks the health of a container using a gRPC service.

Used in: Probe

int32 port = 1
Port number of the gRPC service. Number must be in the range 1 to 65535.
string service = 2
Service is the name of the service to place in the gRPC HealthCheckRequest (see https://github.com/grpc/grpc/blob/master/doc/health-checking.md). If this is not specified, the default behavior is defined by gRPC.

HttpGetAction describes an action based on HTTP Get requests.

Used in: Probe

string path = 1
Path to access on the HTTP server.
int32 port = 2
Number of the port to access on the container. Number must be in the range 1 to 65535.
string host = 3
Host name to connect to, defaults to the model serving container's IP. You probably want to set "Host" in httpHeaders instead.
string scheme = 4
Scheme to use for connecting to the host. Defaults to HTTP. Acceptable values are "HTTP" or "HTTPS".
repeated HttpHeader http_headers = 5
Custom headers to set in the request. HTTP allows repeated headers.

HttpHeader describes a custom header to be used in HTTP probes

Used in: HttpGetAction

string name = 1
The header field name. This will be canonicalized upon output, so case-variant names will be understood as the same header.
string value = 2
The header field value

TcpSocketAction probes the health of a container by opening a TCP socket connection.

Used in: Probe

int32 port = 1
Number of the port to access on the container. Number must be in the range 1 to 65535.
string host = 2
Optional: Host name to connect to, defaults to the model serving container's IP.

PscAutomatedEndpoints defines the output of the forwarding rule automatically created by each PscAutomationConfig.

Used in: IndexPrivateEndpoints

string project_id = 1
Corresponding project_id in pscAutomationConfigs
string network = 2
Corresponding network in pscAutomationConfigs.
string match_address = 3
Ip Address created by the automated forwarding rule.

Actions could take on this Publisher Model.

Used in: PublisherModel

optional CallToAction.ViewRestApi view_rest_api = 1
Optional. To view Rest API docs.
optional CallToAction.RegionalResourceReferences open_notebook = 2
Optional. Open notebook of the PublisherModel.
optional CallToAction.OpenNotebooks open_notebooks = 12
Optional. Open notebooks of the PublisherModel.
optional CallToAction.RegionalResourceReferences create_application = 3
Optional. Create application using the PublisherModel.
optional CallToAction.RegionalResourceReferences open_fine_tuning_pipeline = 4
Optional. Open fine-tuning pipeline of the PublisherModel.
optional CallToAction.OpenFineTuningPipelines open_fine_tuning_pipelines = 13
Optional. Open fine-tuning pipelines of the PublisherModel.
optional CallToAction.RegionalResourceReferences open_prompt_tuning_pipeline = 5
Optional. Open prompt-tuning pipeline of the PublisherModel.
optional CallToAction.RegionalResourceReferences open_genie = 6
Optional. Open Genie / Playground.
optional CallToAction.Deploy deploy = 7
Optional. Deploy the PublisherModel to Vertex Endpoint.
optional CallToAction.DeployGke deploy_gke = 14
Optional. Deploy PublisherModel to Google Kubernetes Engine.
optional CallToAction.RegionalResourceReferences open_generation_ai_studio = 8
Optional. Open in Generation AI Studio.
optional CallToAction.RegionalResourceReferences request_access = 9
Optional. Request for access.
optional CallToAction.RegionalResourceReferences open_evaluation_pipeline = 11
Optional. Open evaluation pipeline of the PublisherModel.

Model metadata that is needed for UploadModel or DeployModel/CreateEndpoint requests.

Used in: CallToAction

oneof prediction_resources
The prediction (for example, the machine) resources that the DeployedModel uses.
- DedicatedResources dedicated_resources = 5
  A description of resources that are dedicated to the DeployedModel, and that need a higher degree of manual configuration.
- AutomaticResources automatic_resources = 6
  A description of resources that to large degree are decided by Vertex AI, and require only a modest additional configuration.
- string shared_resources = 7
  The resource name of the shared DeploymentResourcePool to deploy on. Format: `projects/{project}/locations/{location}/deploymentResourcePools/{deployment_resource_pool}`
string model_display_name = 1
Optional. Default model display name.
optional LargeModelReference large_model_reference = 2
Optional. Large model reference. When this is set, model_artifact_spec is not needed.
optional ModelContainerSpec container_spec = 3
Optional. The specification of the container that is to be used when deploying this Model in Vertex AI. Not present for Large Models.
string artifact_uri = 4
Optional. The path to the directory containing the Model artifact and any of its supporting files.
optional string deploy_task_name = 10
Optional. The name of the deploy task (e.g., "text to image generation").
optional Deploy.DeployMetadata deploy_metadata = 11
Optional. Metadata information about this deployment config.
string title = 8
Required. The title of the regional resource reference.
string public_artifact_uri = 9
Optional. The signed URI for ephemeral Cloud Storage access to model artifact.

Metadata information about the deployment for managing deployment config.

Used in: Deploy

map<string, string> labels = 1
Optional. Labels for the deployment. For managing deployment config like verifying, source of deployment config, etc.
string sample_request = 2
Optional. Sample request for deployed endpoint.

Configurations for PublisherModel GKE deployment

Used in: CallToAction

repeated string gke_yaml_configs = 1
Optional. GKE deployment configuration in yaml format.

Open fine tuning pipelines.

Used in: CallToAction

repeated RegionalResourceReferences fine_tuning_pipelines = 1
Required. Regional resource references to fine tuning pipelines.

Open notebooks.

Used in: CallToAction

repeated RegionalResourceReferences notebooks = 1
Required. Regional resource references to notebooks.

The regional resource name or the URI. Key is region, e.g., us-central1, europe-west2, global, etc..

Used in: CallToAction, OpenFineTuningPipelines, OpenNotebooks

map<string, ResourceReference> references = 1
Required.
string title = 2
Required.
optional string resource_title = 3
Optional. Title of the resource.
optional string resource_use_case = 4
Optional. Use case (CUJ) of the resource.
optional string resource_description = 5
Optional. Description of the resource.

Rest API docs.

Used in: CallToAction

repeated Documentation documentations = 1
Required.
string title = 2
Required. The title of the view rest API.

A named piece of documentation.

Used in: CallToAction.ViewRestApi

string title = 1
Required. E.g., OVERVIEW, USE CASES, DOCUMENTATION, SDK & SAMPLES, JAVA, NODE.JS, etc..
string content = 2
Required. Content of this piece of document (in Markdown format).

An enum representing the launch stage of a PublisherModel.

Used in: PublisherModel

LAUNCH_STAGE_UNSPECIFIED = 0
The model launch stage is unspecified.
EXPERIMENTAL = 1
Used to indicate the PublisherModel is at Experimental launch stage, available to a small set of customers.
PRIVATE_PREVIEW = 2
Used to indicate the PublisherModel is at Private Preview launch stage, only available to a small set of customers, although a larger set of customers than an Experimental launch. Previews are the first launch stage used to get feedback from customers.
PUBLIC_PREVIEW = 3
Used to indicate the PublisherModel is at Public Preview launch stage, available to all customers, although not supported for production workloads.
GA = 4
Used to indicate the PublisherModel is at GA launch stage, available to all customers and ready for production workload.

An enum representing the open source category of a PublisherModel.

Used in: PublisherModel

OPEN_SOURCE_CATEGORY_UNSPECIFIED = 0
The open source category is unspecified, which should not be used.
PROPRIETARY = 1
Used to indicate the PublisherModel is not open sourced.
GOOGLE_OWNED_OSS_WITH_GOOGLE_CHECKPOINT = 2
Used to indicate the PublisherModel is a Google-owned open source model w/ Google checkpoint.
THIRD_PARTY_OWNED_OSS_WITH_GOOGLE_CHECKPOINT = 3
Used to indicate the PublisherModel is a 3p-owned open source model w/ Google checkpoint.
GOOGLE_OWNED_OSS = 4
Used to indicate the PublisherModel is a Google-owned pure open source model.
THIRD_PARTY_OWNED_OSS = 5
Used to indicate the PublisherModel is a 3p-owned pure open source model.

Reference to a resource.

Used in: CallToAction.RegionalResourceReferences

oneof reference
- string uri = 1
  The URI of the resource.
- string resource_name = 2
  The resource name of the Google Cloud resource.
- string use_case = 3
  Use case (CUJ) of the resource.
- string description = 4
  Description of the resource.

An enum representing the state of the PublicModelVersion.

Used in: PublisherModel

VERSION_STATE_UNSPECIFIED = 0
The version state is unspecified.
VERSION_STATE_STABLE = 1
Used to indicate the version is stable.
VERSION_STATE_UNSTABLE = 2
Used to indicate the version is unstable.

View enumeration of PublisherModel.

Used in: GetPublisherModelRequest

PUBLISHER_MODEL_VIEW_UNSPECIFIED = 0
The default / unset value. The API will default to the BASIC view.
PUBLISHER_MODEL_VIEW_BASIC = 1
Include basic metadata about the publisher model, but not the full contents.
PUBLISHER_MODEL_VIEW_FULL = 2
Include everything.
PUBLISHER_MODEL_VERSION_VIEW_BASIC = 3
Include: VersionId, ModelVersionExternalName, and SupportedActions.

Details of operations that perform [MetadataService.PurgeArtifacts][google.cloud.aiplatform.v1.MetadataService.PurgeArtifacts].

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for purging Artifacts.

Response message for [MetadataService.PurgeArtifacts][google.cloud.aiplatform.v1.MetadataService.PurgeArtifacts].

int64 purge_count = 1
The number of Artifacts that this request deleted (or, if `force` is false, the number of Artifacts that will be deleted). This can be an estimate.
repeated string purge_sample = 2
A sample of the Artifact names that will be deleted. Only populated if `force` is set to false. The maximum number of samples is 100 (it is possible to return fewer).

Details of operations that perform [MetadataService.PurgeContexts][google.cloud.aiplatform.v1.MetadataService.PurgeContexts].

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for purging Contexts.

Response message for [MetadataService.PurgeContexts][google.cloud.aiplatform.v1.MetadataService.PurgeContexts].

int64 purge_count = 1
The number of Contexts that this request deleted (or, if `force` is false, the number of Contexts that will be deleted). This can be an estimate.
repeated string purge_sample = 2
A sample of the Context names that will be deleted. Only populated if `force` is set to false. The maximum number of samples is 100 (it is possible to return fewer).

Details of operations that perform [MetadataService.PurgeExecutions][google.cloud.aiplatform.v1.MetadataService.PurgeExecutions].

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for purging Executions.

Response message for [MetadataService.PurgeExecutions][google.cloud.aiplatform.v1.MetadataService.PurgeExecutions].

int64 purge_count = 1
The number of Executions that this request deleted (or, if `force` is false, the number of Executions that will be deleted). This can be an estimate.
repeated string purge_sample = 2
A sample of the Execution names that will be deleted. Only populated if `force` is set to false. The maximum number of samples is 100 (it is possible to return fewer).

The spec of a Python packaged code.

Used in: WorkerPoolSpec

string executor_image_uri = 1
Required. The URI of a container image in Artifact Registry that will run the provided Python package. Vertex AI provides a wide range of executor images with pre-installed packages to meet users' various use cases. See the list of [pre-built containers for training](https://cloud.google.com/vertex-ai/docs/training/pre-built-containers). You must use an image from this list.
repeated string package_uris = 2
Required. The Google Cloud Storage location of the Python package files which are the training program and its dependent packages. The maximum number of package URIs is 100.
string python_module = 3
Required. The Python module name to run after installing the packages.
repeated string args = 4
Command line arguments to be passed to the Python task.
repeated EnvVar env = 5
Environment variables to be passed to the python module. Maximum limit is 100.

Input for question answering correctness metric.

optional QuestionAnsweringCorrectnessSpec metric_spec = 1
Required. Spec for question answering correctness score metric.
optional QuestionAnsweringCorrectnessInstance instance = 2
Required. Question answering correctness instance.

Spec for question answering correctness instance.

Used in: QuestionAnsweringCorrectnessInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Optional. Ground truth used to compare against the prediction.
optional string context = 3
Optional. Text provided as context to answer the question.
optional string instruction = 4
Required. The question asked and other instruction in the inference prompt.

Spec for question answering correctness result.

optional float score = 1
Output only. Question Answering Correctness score.
string explanation = 2
Output only. Explanation for question answering correctness score.
optional float confidence = 3
Output only. Confidence for question answering correctness score.

Spec for question answering correctness metric.

Used in: QuestionAnsweringCorrectnessInput

bool use_reference = 1
Optional. Whether to use instance.reference to compute question answering correctness.
int32 version = 2
Optional. Which version to use for evaluation.

Input for question answering helpfulness metric.

optional QuestionAnsweringHelpfulnessSpec metric_spec = 1
Required. Spec for question answering helpfulness score metric.
optional QuestionAnsweringHelpfulnessInstance instance = 2
Required. Question answering helpfulness instance.

Spec for question answering helpfulness instance.

Used in: QuestionAnsweringHelpfulnessInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Optional. Ground truth used to compare against the prediction.
optional string context = 3
Optional. Text provided as context to answer the question.
optional string instruction = 4
Required. The question asked and other instruction in the inference prompt.

Spec for question answering helpfulness result.

optional float score = 1
Output only. Question Answering Helpfulness score.
string explanation = 2
Output only. Explanation for question answering helpfulness score.
optional float confidence = 3
Output only. Confidence for question answering helpfulness score.

Spec for question answering helpfulness metric.

Used in: QuestionAnsweringHelpfulnessInput

bool use_reference = 1
Optional. Whether to use instance.reference to compute question answering helpfulness.
int32 version = 2
Optional. Which version to use for evaluation.

Input for question answering quality metric.

optional QuestionAnsweringQualitySpec metric_spec = 1
Required. Spec for question answering quality score metric.
optional QuestionAnsweringQualityInstance instance = 2
Required. Question answering quality instance.

Spec for question answering quality instance.

Used in: QuestionAnsweringQualityInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Optional. Ground truth used to compare against the prediction.
optional string context = 3
Required. Text to answer the question.
optional string instruction = 4
Required. Question Answering prompt for LLM.

Spec for question answering quality result.

optional float score = 1
Output only. Question Answering Quality score.
string explanation = 2
Output only. Explanation for question answering quality score.
optional float confidence = 3
Output only. Confidence for question answering quality score.

Spec for question answering quality score metric.

Used in: QuestionAnsweringQualityInput

bool use_reference = 1
Optional. Whether to use instance.reference to compute question answering quality.
int32 version = 2
Optional. Which version to use for evaluation.

Input for question answering relevance metric.

optional QuestionAnsweringRelevanceSpec metric_spec = 1
Required. Spec for question answering relevance score metric.
optional QuestionAnsweringRelevanceInstance instance = 2
Required. Question answering relevance instance.

Spec for question answering relevance instance.

Used in: QuestionAnsweringRelevanceInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Optional. Ground truth used to compare against the prediction.
optional string context = 3
Optional. Text provided as context to answer the question.
optional string instruction = 4
Required. The question asked and other instruction in the inference prompt.

Spec for question answering relevance result.

optional float score = 1
Output only. Question Answering Relevance score.
string explanation = 2
Output only. Explanation for question answering relevance score.
optional float confidence = 3
Output only. Confidence for question answering relevance score.

Spec for question answering relevance metric.

Used in: QuestionAnsweringRelevanceInput

bool use_reference = 1
Optional. Whether to use instance.reference to compute question answering relevance.
int32 version = 2
Optional. Which version to use for evaluation.

A RagChunk includes the content of a chunk of a RagFile, and associated metadata.

Used in: Fact, GroundingChunk.RetrievedContext, RagContexts.Context

string text = 1
The content of the chunk.
optional RagChunk.PageSpan page_span = 2
If populated, represents where the chunk starts and ends in the document.

Represents where the chunk starts and ends in the document.

Used in: RagChunk

int32 first_page = 1
Page where chunk starts in the document. Inclusive. 1-indexed.
int32 last_page = 2
Page where chunk ends in the document. Inclusive. 1-indexed.

Relevant contexts for one query.

Used in: RetrieveContextsResponse

repeated RagContexts.Context contexts = 1
All its contexts.

A context of the query.

Used in: RagContexts

string source_uri = 1
If the file is imported from Cloud Storage or Google Drive, source_uri will be original file URI in Cloud Storage or Google Drive; if file is uploaded, source_uri will be file display name.
string source_display_name = 5
The file display name.
string text = 2
The text chunk.
optional double score = 6
According to the underlying Vector DB and the selected metric type, the score can be either the distance or the similarity between the query and the context and its range depends on the metric type. For example, if the metric type is COSINE_DISTANCE, it represents the distance between the query and the context. The larger the distance, the less relevant the context is to the query. The range is [0, 2], while 0 means the most relevant and 2 means the least relevant.
optional RagChunk chunk = 7
Context of the retrieved chunk.

A RagCorpus is a RagFile container and a project can have multiple RagCorpora.

Used as response type in: VertexRagDataService.GetRagCorpus

Used as field type in: CreateRagCorpusRequest, ListRagCorporaResponse, UpdateRagCorpusRequest

oneof backend_config
The backend config of the RagCorpus. It can be data store and/or retrieval engine.
- RagVectorDbConfig vector_db_config = 9
  Optional. Immutable. The config for the Vector DBs.
- VertexAiSearchConfig vertex_ai_search_config = 10
  Optional. Immutable. The config for the Vertex AI Search.
string name = 1
Output only. The resource name of the RagCorpus.
string display_name = 2
Required. The display name of the RagCorpus. The name can be up to 128 characters long and can consist of any UTF-8 characters.
string description = 3
Optional. The description of the RagCorpus.
optional protobuf.Timestamp create_time = 4
Output only. Timestamp when this RagCorpus was created.
optional protobuf.Timestamp update_time = 5
Output only. Timestamp when this RagCorpus was last updated.
optional CorpusStatus corpus_status = 8
Output only. RagCorpus state.

Config for the embedding model to use for RAG.

Used in: RagVectorDbConfig

oneof model_config
The model config to use.
- RagEmbeddingModelConfig.VertexPredictionEndpoint vertex_prediction_endpoint = 1
  The Vertex AI Prediction Endpoint that either refers to a publisher model or an endpoint that is hosting a 1P fine-tuned text embedding model. Endpoints hosting non-1P fine-tuned text embedding models are currently not supported. This is used for dense vector search.

Config representing a model hosted on Vertex Prediction Endpoint.

Used in: RagEmbeddingModelConfig

string endpoint = 1
Required. The endpoint resource name. Format: `projects/{project}/locations/{location}/publishers/{publisher}/models/{model}` or `projects/{project}/locations/{location}/endpoints/{endpoint}`
string model = 2
Output only. The resource name of the model that is deployed on the endpoint. Present only when the endpoint is not a publisher model. Pattern: `projects/{project}/locations/{location}/models/{model}`
string model_version_id = 3
Output only. Version ID of the model that is deployed on the endpoint. Present only when the endpoint is not a publisher model.

A RagFile contains user data for chunking, embedding and indexing.

Used as response type in: VertexRagDataService.GetRagFile

Used as field type in: ListRagFilesResponse, UploadRagFileRequest, UploadRagFileResponse

oneof rag_file_source
The origin location of the RagFile if it is imported from Google Cloud Storage or Google Drive.
- GcsSource gcs_source = 8
  Output only. Google Cloud Storage location of the RagFile. It does not support wildcards in the Cloud Storage uri for now.
- GoogleDriveSource google_drive_source = 9
  Output only. Google Drive location. Supports importing individual files as well as Google Drive folders.
- DirectUploadSource direct_upload_source = 10
  Output only. The RagFile is encapsulated and uploaded in the UploadRagFile request.
- SlackSource slack_source = 11
  The RagFile is imported from a Slack channel.
- JiraSource jira_source = 12
  The RagFile is imported from a Jira query.
- SharePointSources share_point_sources = 14
  The RagFile is imported from a SharePoint source.
string name = 1
Output only. The resource name of the RagFile.
string display_name = 2
Required. The display name of the RagFile. The name can be up to 128 characters long and can consist of any UTF-8 characters.
string description = 3
Optional. The description of the RagFile.
optional protobuf.Timestamp create_time = 6
Output only. Timestamp when this RagFile was created.
optional protobuf.Timestamp update_time = 7
Output only. Timestamp when this RagFile was last updated.
optional FileStatus file_status = 13
Output only. State of the RagFile.

Specifies the size and overlap of chunks for RagFiles.

Used in: RagFileTransformationConfig

oneof chunking_config
Specifies the chunking config for RagFiles.
- RagFileChunkingConfig.FixedLengthChunking fixed_length_chunking = 3
  Specifies the fixed length chunking config.

Specifies the fixed length chunking config.

Used in: RagFileChunkingConfig

int32 chunk_size = 1
The size of the chunks.
int32 chunk_overlap = 2
The overlap between chunks.

Specifies the parsing config for RagFiles.

Used in: ImportRagFilesConfig

oneof parser
The parser to use for RagFiles.
- RagFileParsingConfig.LayoutParser layout_parser = 4
  The Layout Parser to use for RagFiles.
- RagFileParsingConfig.LlmParser llm_parser = 5
  The LLM Parser to use for RagFiles.

Document AI Layout Parser config.

Used in: RagFileParsingConfig

string processor_name = 1
The full resource name of a Document AI processor or processor version. The processor must have type `LAYOUT_PARSER_PROCESSOR`. If specified, the `additional_config.parse_as_scanned_pdf` field must be false. Format: * `projects/{project_id}/locations/{location}/processors/{processor_id}` * `projects/{project_id}/locations/{location}/processors/{processor_id}/processorVersions/{processor_version_id}`
int32 max_parsing_requests_per_min = 2
The maximum number of requests the job is allowed to make to the Document AI processor per minute. Consult https://cloud.google.com/document-ai/quotas and the Quota page for your project to set an appropriate value here. If unspecified, a default value of 120 QPM would be used.

Specifies the advanced parsing for RagFiles.

Used in: RagFileParsingConfig

string model_name = 1
The name of a LLM model used for parsing. Format: * `projects/{project_id}/locations/{location}/publishers/{publisher}/models/{model}`
int32 max_parsing_requests_per_min = 2
The maximum number of requests the job is allowed to make to the LLM model per minute. Consult https://cloud.google.com/vertex-ai/generative-ai/docs/quotas and your document size to set an appropriate value here. If unspecified, a default value of 5000 QPM would be used.
string custom_parsing_prompt = 3
The prompt to use for parsing. If not specified, a default prompt will be used.

Specifies the transformation config for RagFiles.

Used in: ImportRagFilesConfig, UploadRagFileConfig

optional RagFileChunkingConfig rag_file_chunking_config = 1
Specifies the chunking config for RagFiles.

A query to retrieve relevant contexts.

Used in: RetrieveContextsRequest

oneof query
The query to retrieve contexts. Currently only text query is supported.
- string text = 1
  Optional. The query in text format to get relevant contexts.
optional RagRetrievalConfig rag_retrieval_config = 6
Optional. The retrieval config for the query.

Specifies the context retrieval config.

Used in: RagQuery, VertexRagStore

int32 top_k = 1
Optional. The number of contexts to retrieve.
optional RagRetrievalConfig.Filter filter = 3
Optional. Config for filters.
optional RagRetrievalConfig.Ranking ranking = 4
Optional. Config for ranking and reranking.

Config for filters.

Used in: RagRetrievalConfig

oneof vector_db_threshold
Filter contexts retrieved from the vector DB based on either vector distance or vector similarity.
- double vector_distance_threshold = 3
  Optional. Only returns contexts with vector distance smaller than the threshold.
- double vector_similarity_threshold = 4
  Optional. Only returns contexts with vector similarity larger than the threshold.
string metadata_filter = 2
Optional. String for metadata filtering.

Config for ranking and reranking.

Used in: RagRetrievalConfig

oneof ranking_config
Config options for ranking. Currently only Rank Service is supported.
- Ranking.RankService rank_service = 1
  Optional. Config for Rank Service.
- Ranking.LlmRanker llm_ranker = 3
  Optional. Config for LlmRanker.

Config for LlmRanker.

Used in: Ranking

optional string model_name = 1
Optional. The model name used for ranking. Format: `gemini-1.5-pro`

Config for Rank Service.

Used in: Ranking

optional string model_name = 1
Optional. The model name of the rank service. Format: `semantic-ranker-512@latest`

Config for the Vector DB to use for RAG.

Used in: RagCorpus

oneof vector_db
The config for the Vector DB.
- RagVectorDbConfig.RagManagedDb rag_managed_db = 1
  The config for the RAG-managed Vector DB.
- RagVectorDbConfig.Pinecone pinecone = 3
  The config for the Pinecone.
- RagVectorDbConfig.VertexVectorSearch vertex_vector_search = 6
  The config for the Vertex Vector Search.
optional ApiAuth api_auth = 5
Authentication config for the chosen Vector DB.
optional RagEmbeddingModelConfig rag_embedding_model_config = 7
Optional. Immutable. The embedding model config of the Vector DB.

The config for the Pinecone.

Used in: RagVectorDbConfig

string index_name = 1
Pinecone index name. This value cannot be changed after it's set.

The config for the default RAG-managed Vector DB.

Used in: RagVectorDbConfig

(message has no fields)

The config for the Vertex Vector Search.

Used in: RagVectorDbConfig

string index_endpoint = 1
The resource name of the Index Endpoint. Format: `projects/{project}/locations/{location}/indexEndpoints/{index_endpoint}`
string index = 2
The resource name of the Index. Format: `projects/{project}/locations/{location}/indexes/{index}`

Configuration for the Ray OSS Logs.

Used in: RaySpec

bool disabled = 1
Optional. Flag to disable the export of Ray OSS logs to Cloud Logging.

Configuration for the Ray metrics.

Used in: RaySpec

bool disabled = 1
Optional. Flag to disable the Ray metrics collection.

Configuration information for the Ray cluster. For experimental launch, Ray cluster creation and Persistent cluster creation are 1:1 mapping: We will provision all the nodes within the Persistent cluster as Ray nodes.

Used in: ResourceRuntimeSpec

string image_uri = 1
Optional. Default image for user to choose a preferred ML framework (for example, TensorFlow or Pytorch) by choosing from [Vertex prebuilt images](https://cloud.google.com/vertex-ai/docs/training/pre-built-containers). Either this or the resource_pool_images is required. Use this field if you need all the resource pools to have the same Ray image. Otherwise, use the {@code resource_pool_images} field.
map<string, string> resource_pool_images = 6
Optional. Required if image_uri isn't set. A map of resource_pool_id to prebuild Ray image if user need to use different images for different head/worker pools. This map needs to cover all the resource pool ids. Example: { "ray_head_node_pool": "head image" "ray_worker_node_pool1": "worker image" "ray_worker_node_pool2": "another worker image" }
string head_node_resource_pool_id = 7
Optional. This will be used to indicate which resource pool will serve as the Ray head node(the first node within that pool). Will use the machine from the first workerpool as the head node by default if this field isn't set.
optional RayMetricSpec ray_metric_spec = 8
Optional. Ray metrics configurations.
optional RayLogsSpec ray_logs_spec = 10
Optional. OSS Ray logging configurations.

Response message for [FeaturestoreOnlineServingService.ReadFeatureValues][google.cloud.aiplatform.v1.FeaturestoreOnlineServingService.ReadFeatureValues].

Used as response type in: FeaturestoreOnlineServingService.ReadFeatureValues, FeaturestoreOnlineServingService.StreamingReadFeatureValues

optional ReadFeatureValuesResponse.Header header = 1
Response header.
optional ReadFeatureValuesResponse.EntityView entity_view = 2
Entity view with Feature values. This may be the entity in the Featurestore if values for all Features were requested, or a projection of the entity in the Featurestore if values for only some Features were requested.

Entity view with Feature values.

Used in: ReadFeatureValuesResponse

string entity_id = 1
ID of the requested entity.
repeated EntityView.Data data = 2
Each piece of data holds the k requested values for one requested Feature. If no values for the requested Feature exist, the corresponding cell will be empty. This has the same size and is in the same order as the features from the header [ReadFeatureValuesResponse.header][google.cloud.aiplatform.v1.ReadFeatureValuesResponse.header].

Container to hold value(s), successive in time, for one Feature from the request.

Used in: EntityView

oneof data
- FeatureValue value = 1
  Feature value if a single value is requested.
- FeatureValueList values = 2
  Feature values list if values, successive in time, are requested. If the requested number of values is greater than the number of existing Feature values, nonexistent values are omitted instead of being returned as empty.

Metadata for requested Features.

Used in: Header

string id = 1
Feature ID.

Response header with metadata for the requested [ReadFeatureValuesRequest.entity_type][google.cloud.aiplatform.v1.ReadFeatureValuesRequest.entity_type] and Features.

Used in: ReadFeatureValuesResponse

string entity_type = 1
The resource name of the EntityType from the [ReadFeatureValuesRequest][google.cloud.aiplatform.v1.ReadFeatureValuesRequest]. Value format: `projects/{project}/locations/{location}/featurestores/{featurestore}/entityTypes/{entityType}`.
repeated FeatureDescriptor feature_descriptors = 2
List of Feature metadata corresponding to each piece of [ReadFeatureValuesResponse.EntityView.data][google.cloud.aiplatform.v1.ReadFeatureValuesResponse.EntityView.data].

Per month usage data

Used in: ReadTensorboardUsageResponse

repeated PerUserUsageData user_usage_data = 1
Usage data for each user in the given month.

Per user usage data.

Used in: PerMonthUsageData

string username = 1
User's username
int64 view_count = 2
Number of times the user has read data within the Tensorboard.

ReasoningEngine provides a customizable runtime for models to determine which actions to take and in which order.

Used as response type in: ReasoningEngineService.GetReasoningEngine

Used as field type in: CreateReasoningEngineRequest, ListReasoningEnginesResponse, UpdateReasoningEngineRequest

string name = 1
Identifier. The resource name of the ReasoningEngine.
string display_name = 2
Required. The display name of the ReasoningEngine.
string description = 7
Optional. The description of the ReasoningEngine.
optional ReasoningEngineSpec spec = 3
Optional. Configurations of the ReasoningEngine
optional protobuf.Timestamp create_time = 4
Output only. Timestamp when this ReasoningEngine was created.
optional protobuf.Timestamp update_time = 5
Output only. Timestamp when this ReasoningEngine was most recently updated.
string etag = 6
Optional. Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.

ReasoningEngine configurations

Used in: ReasoningEngine

optional ReasoningEngineSpec.PackageSpec package_spec = 2
Optional. User provided package spec of the ReasoningEngine. Ignored when users directly specify a deployment image through `deployment_spec.first_party_image_override`, but keeping the field_behavior to avoid introducing breaking changes.
optional ReasoningEngineSpec.DeploymentSpec deployment_spec = 4
Optional. The specification of a Reasoning Engine deployment.
repeated protobuf.Struct class_methods = 3
Optional. Declarations for object class methods in OpenAPI specification format.
string agent_framework = 5
Optional. The OSS agent framework used to develop the agent. Currently supported values: "langchain", "langgraph", "ag2", "custom".

The specification of a Reasoning Engine deployment.

Used in: ReasoningEngineSpec

repeated EnvVar env = 1
Optional. Environment variables to be set with the Reasoning Engine deployment. The environment variables can be updated through the UpdateReasoningEngine API.
repeated SecretEnvVar secret_env = 2
Optional. Environment variables where the value is a secret in Cloud Secret Manager. To use this feature, add 'Secret Manager Secret Accessor' role (roles/secretmanager.secretAccessor) to AI Platform Reasoning Engine Service Agent.

User provided package spec like pickled object and package requirements.

Used in: ReasoningEngineSpec

string pickle_object_gcs_uri = 1
Optional. The Cloud Storage URI of the pickled python object.
string dependency_files_gcs_uri = 2
Optional. The Cloud Storage URI of the dependency files in tar.gz format.
string requirements_gcs_uri = 3
Optional. The Cloud Storage URI of the `requirements.txt` file
string python_version = 4
Optional. The Python version. Currently support 3.8, 3.9, 3.10, 3.11. If not specified, default value is 3.10.

Runtime operation information for [GenAiTuningService.RebaseTunedModel][google.cloud.aiplatform.v1.GenAiTuningService.RebaseTunedModel].

optional GenericOperationMetadata generic_metadata = 1
The common part of the operation generic information.

Details of operations that perform reboot PersistentResource.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for PersistentResource.
string progress_message = 2
Progress Message for Reboot LRO

A ReservationAffinity can be used to configure a Vertex AI resource (e.g., a DeployedModel) to draw its Compute Engine resources from a Shared Reservation, or exclusively from on-demand capacity.

Used in: MachineSpec

ReservationAffinity.Type reservation_affinity_type = 1
Required. Specifies the reservation affinity type.
string key = 2
Optional. Corresponds to the label key of a reservation resource. To target a SPECIFIC_RESERVATION by name, use `compute.googleapis.com/reservation-name` as the key and specify the name of your reservation as its value.
repeated string values = 3
Optional. Corresponds to the label values of a reservation resource. This must be the full resource name of the reservation.

Identifies a type of reservation affinity.

Used in: ReservationAffinity

TYPE_UNSPECIFIED = 0
Default value. This should not be used.
NO_RESERVATION = 1
Do not consume from any reserved capacity, only use on-demand.
ANY_RESERVATION = 2
Consume any reservation available, falling back to on-demand.
SPECIFIC_RESERVATION = 3
Consume from a specific reservation. When chosen, the reservation must be identified via the `key` and `values` fields.

Represents the spec of a group of resources of the same type, for example machine type, disk, and accelerators, in a PersistentResource.

Used in: PersistentResource

string id = 1
Immutable. The unique ID in a PersistentResource for referring to this resource pool. User can specify it if necessary. Otherwise, it's generated automatically.
optional MachineSpec machine_spec = 2
Required. Immutable. The specification of a single machine.
optional int64 replica_count = 3
Optional. The total number of machines to use for this resource pool.
optional DiskSpec disk_spec = 4
Optional. Disk spec for the machine in this node pool.
int64 used_replica_count = 6
Output only. The number of machines currently in use by training jobs for this resource pool. Will replace idle_replica_count.
optional ResourcePool.AutoscalingSpec autoscaling_spec = 7
Optional. Optional spec to configure GKE or Ray-on-Vertex autoscaling

The min/max number of replicas allowed if enabling autoscaling

Used in: ResourcePool

optional int64 min_replica_count = 1
Optional. min replicas in the node pool, must be ≤ replica_count and < max_replica_count or will throw error. For autoscaling enabled Ray-on-Vertex, we allow min_replica_count of a resource_pool to be 0 to match the OSS Ray behavior(https://docs.ray.io/en/latest/cluster/vms/user-guides/configuring-autoscaling.html#cluster-config-parameters). As for Persistent Resource, the min_replica_count must be > 0, we added a corresponding validation inside CreatePersistentResourceRequestValidator.java.
optional int64 max_replica_count = 2
Optional. max replicas in the node pool, must be ≥ replica_count and > min_replica_count or will throw error

Persistent Cluster runtime information as output

Used in: PersistentResource

map<string, string> access_uris = 1
Output only. URIs for user to connect to the Cluster. Example: { "RAY_HEAD_NODE_INTERNAL_IP": "head-node-IP:10001" "RAY_DASHBOARD_URI": "ray-dashboard-address:8888" }

Configuration for the runtime on a PersistentResource instance, including but not limited to: * Service accounts used to run the workloads. * Whether to make it a dedicated Ray Cluster.

Used in: PersistentResource

optional ServiceAccountSpec service_account_spec = 2
Optional. Configure the use of workload identity on the PersistentResource
optional RaySpec ray_spec = 1
Optional. Ray cluster configuration. Required when creating a dedicated RayCluster on the PersistentResource.

Statistics information about resource consumption.

Used in: BatchPredictionJob

double replica_hours = 1
Output only. The number of replica hours used. Note that many replicas may run in parallel, and additionally any given work may be queued for some time. Therefore this value is not strictly related to wall time.

Runtime operation information for [DatasetService.RestoreDatasetVersion][google.cloud.aiplatform.v1.DatasetService.RestoreDatasetVersion].

optional GenericOperationMetadata generic_metadata = 1
The common part of the operation metadata.

Defines a retrieval tool that model can call to access external knowledge.

Used in: Tool

oneof source
The source of the retrieval.
- VertexAISearch vertex_ai_search = 2
  Set to use data source powered by Vertex AI Search.
- VertexRagStore vertex_rag_store = 4
  Set to use data source powered by Vertex RAG store. User data is uploaded via the VertexRagDataService.
bool disable_attribution = 3
Optional. Deprecated. This option is no longer supported.

Retrieval config.

Used in: ToolConfig

optional type.LatLng lat_lng = 1
The location of the user.
optional string language_code = 2
The language code of the user.

Metadata related to retrieval in the grounding flow.

Used in: GroundingMetadata

float google_search_dynamic_retrieval_score = 2
Optional. Score indicating how likely information from Google Search could help answer the prompt. The score is in the range `[0, 1]`, where 0 is the least likely and 1 is the most likely. This score is only populated when Google Search grounding and dynamic retrieval is enabled. It will be compared to the threshold to determine whether to trigger Google Search.

The data source for Vertex RagStore.

Used in: RetrieveContextsRequest

repeated VertexRagStore.RagResource rag_resources = 3
Optional. The representation of the rag source. It can be used to specify corpus only or ragfiles. Currently only support one corpus or multiple files from one corpus. In the future we may open up multiple corpora support.
optional double vector_distance_threshold = 2
Optional. Only return contexts with vector distance smaller than the threshold.

The definition of the Rag resource.

Used in: VertexRagStore

string rag_corpus = 1
Optional. RagCorpora resource name. Format: `projects/{project}/locations/{location}/ragCorpora/{rag_corpus}`
repeated string rag_file_ids = 2
Optional. rag_file_id. The files should be in the same rag_corpus set in rag_corpus field.

Input for rouge metric.

optional RougeSpec metric_spec = 1
Required. Spec for rouge score metric.
repeated RougeInstance instances = 2
Required. Repeated rouge instances.

Spec for rouge instance.

Used in: RougeInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Required. Ground truth used to compare against the prediction.

Rouge metric value for an instance.

Used in: RougeResults

optional float score = 1
Output only. Rouge score.

Results for rouge metric.

repeated RougeMetricValue rouge_metric_values = 1
Output only. Rouge metric values.

Spec for rouge score metric - calculates the recall of n-grams in prediction as compared to reference - returns a score ranging between 0 and 1.

Used in: RougeInput

string rouge_type = 1
Optional. Supported rouge types are rougen[1-9], rougeL, and rougeLsum.
bool use_stemmer = 2
Optional. Whether to use stemmer to compute rouge score.
bool split_summaries = 3
Optional. Whether to split summaries while using rougeLsum.

Input for safety metric.

optional SafetySpec metric_spec = 1
Required. Spec for safety metric.
optional SafetyInstance instance = 2
Required. Safety instance.

Spec for safety instance.

Used in: SafetyInput

optional string prediction = 1
Required. Output of the evaluated model.

Safety rating corresponding to the generated content.

Used in: Candidate, GenerateContentResponse.PromptFeedback

HarmCategory category = 1
Output only. Harm category.
SafetyRating.HarmProbability probability = 2
Output only. Harm probability levels in the content.
float probability_score = 5
Output only. Harm probability score.
SafetyRating.HarmSeverity severity = 6
Output only. Harm severity levels in the content.
float severity_score = 7
Output only. Harm severity score.
bool blocked = 3
Output only. Indicates whether the content was filtered out because of this rating.

Harm probability levels in the content.

Used in: SafetyRating

HARM_PROBABILITY_UNSPECIFIED = 0
Harm probability unspecified.
NEGLIGIBLE = 1
Negligible level of harm.
LOW = 2
Low level of harm.
MEDIUM = 3
Medium level of harm.
HIGH = 4
High level of harm.

Harm severity levels.

Used in: SafetyRating

HARM_SEVERITY_UNSPECIFIED = 0
Harm severity unspecified.
HARM_SEVERITY_NEGLIGIBLE = 1
Negligible level of harm severity.
HARM_SEVERITY_LOW = 2
Low level of harm severity.
HARM_SEVERITY_MEDIUM = 3
Medium level of harm severity.
HARM_SEVERITY_HIGH = 4
High level of harm severity.

Spec for safety result.

optional float score = 1
Output only. Safety score.
string explanation = 2
Output only. Explanation for safety score.
optional float confidence = 3
Output only. Confidence for safety score.

Safety settings.

Used in: GenerateContentRequest

HarmCategory category = 1
Required. Harm category.
SafetySetting.HarmBlockThreshold threshold = 2
Required. The harm block threshold.
SafetySetting.HarmBlockMethod method = 4
Optional. Specify if the threshold is used for probability or severity score. If not specified, the threshold is used for probability score.

Probability vs severity.

Used in: SafetySetting

HARM_BLOCK_METHOD_UNSPECIFIED = 0
The harm block method is unspecified.
SEVERITY = 1
The harm block method uses both probability and severity scores.
PROBABILITY = 2
The harm block method uses the probability score.

Probability based thresholds levels for blocking.

Used in: SafetySetting

HARM_BLOCK_THRESHOLD_UNSPECIFIED = 0
Unspecified harm block threshold.
BLOCK_LOW_AND_ABOVE = 1
Block low threshold and above (i.e. block more).
BLOCK_MEDIUM_AND_ABOVE = 2
Block medium threshold and above.
BLOCK_ONLY_HIGH = 3
Block only high threshold (i.e. block less).
BLOCK_NONE = 4
Block none.
OFF = 5
Turn off the safety filter.

Spec for safety metric.

Used in: SafetyInput

int32 version = 1
Optional. Which version to use for evaluation.

Active learning data sampling config. For every active learning labeling iteration, it will select a batch of data based on the sampling strategy.

Used in: ActiveLearningConfig

oneof initial_batch_sample_size
Decides sample size for the initial batch. initial_batch_sample_percentage is used by default.
- int32 initial_batch_sample_percentage = 1
  The percentage of data needed to be labeled in the first batch.
oneof following_batch_sample_size
Decides sample size for the following batches. following_batch_sample_percentage is used by default.
- int32 following_batch_sample_percentage = 3
  The percentage of data needed to be labeled in each following batch (except the first batch).
SampleConfig.SampleStrategy sample_strategy = 5
Field to choose sampling strategy. Sampling strategy will decide which data should be selected for human labeling in every batch.

Sample strategy decides which subset of DataItems should be selected for human labeling in every batch.

Used in: SampleConfig

SAMPLE_STRATEGY_UNSPECIFIED = 0
Default will be treated as UNCERTAINTY.
UNCERTAINTY = 1
Sample the most uncertain data to label.

An attribution method that approximates Shapley values for features that contribute to the label being predicted. A sampling strategy is used to approximate the value rather than considering all subsets of features.

Used in: ExplanationParameters

int32 path_count = 1
Required. The number of feature permutations to consider when approximating the Shapley values. Valid range of its value is [1, 50], inclusively.

Sampling Strategy for logging, can be for both training and prediction dataset.

Used in: ModelDeploymentMonitoringJob, ModelMonitoringObjectiveConfig.TrainingDataset

optional SamplingStrategy.RandomSampleConfig random_sample_config = 1
Random sample config. Will support more sampling strategies later.

Requests are randomly selected.

Used in: SamplingStrategy

double sample_rate = 1
Sample rate (0, 1]

A SavedQuery is a view of the dataset. It references a subset of annotations by problem type and filters.

Used in: Dataset, ListSavedQueriesResponse

string name = 1
Output only. Resource name of the SavedQuery.
string display_name = 2
Required. The user-defined name of the SavedQuery. The name can be up to 128 characters long and can consist of any UTF-8 characters.
optional protobuf.Value metadata = 12
Some additional information about the SavedQuery.
optional protobuf.Timestamp create_time = 3
Output only. Timestamp when this SavedQuery was created.
optional protobuf.Timestamp update_time = 4
Output only. Timestamp when SavedQuery was last updated.
string annotation_filter = 5
Output only. Filters on the Annotations in the dataset.
string problem_type = 6
Required. Problem type of the SavedQuery. Allowed values: * IMAGE_CLASSIFICATION_SINGLE_LABEL * IMAGE_CLASSIFICATION_MULTI_LABEL * IMAGE_BOUNDING_POLY * IMAGE_BOUNDING_BOX * TEXT_CLASSIFICATION_SINGLE_LABEL * TEXT_CLASSIFICATION_MULTI_LABEL * TEXT_EXTRACTION * TEXT_SENTIMENT * VIDEO_CLASSIFICATION * VIDEO_OBJECT_TRACKING
int32 annotation_spec_count = 10
Output only. Number of AnnotationSpecs in the context of the SavedQuery.
string etag = 8
Used to perform a consistent read-modify-write update. If not set, a blind "overwrite" update happens.
bool support_automl_training = 9
Output only. If the Annotations belonging to the SavedQuery can be used for AutoML training.

One point viewable on a scalar metric plot.

Used in: TimeSeriesDataPoint

double value = 1
Value of the point at this step / timestamp.

An instance of a Schedule periodically schedules runs to make API calls based on user specified time specification and API request type.

Used as response type in: ScheduleService.CreateSchedule, ScheduleService.GetSchedule, ScheduleService.UpdateSchedule

Used as field type in: CreateScheduleRequest, ListSchedulesResponse, UpdateScheduleRequest

oneof time_specification
Required. The time specification to launch scheduled runs.
- string cron = 10
  Cron schedule (https://en.wikipedia.org/wiki/Cron) to launch scheduled runs. To explicitly set a timezone to the cron tab, apply a prefix in the cron tab: "CRON_TZ=${IANA_TIME_ZONE}" or "TZ=${IANA_TIME_ZONE}". The ${IANA_TIME_ZONE} may only be a valid string from IANA time zone database. For example, "CRON_TZ=America/New_York 1 * * * *", or "TZ=America/New_York 1 * * * *".
oneof request
Required. The API request template to launch the scheduled runs. User-specified ID is not supported in the request template.
- CreatePipelineJobRequest create_pipeline_job_request = 14
  Request for [PipelineService.CreatePipelineJob][google.cloud.aiplatform.v1.PipelineService.CreatePipelineJob]. CreatePipelineJobRequest.parent field is required (format: projects/{project}/locations/{location}).
- CreateNotebookExecutionJobRequest create_notebook_execution_job_request = 20
  Request for [NotebookService.CreateNotebookExecutionJob][google.cloud.aiplatform.v1.NotebookService.CreateNotebookExecutionJob].
string name = 1
Immutable. The resource name of the Schedule.
string display_name = 2
Required. User provided name of the Schedule. The name can be up to 128 characters long and can consist of any UTF-8 characters.
optional protobuf.Timestamp start_time = 3
Optional. Timestamp after which the first run can be scheduled. Default to Schedule create time if not specified.
optional protobuf.Timestamp end_time = 4
Optional. Timestamp after which no new runs can be scheduled. If specified, The schedule will be completed when either end_time is reached or when scheduled_run_count >= max_run_count. If not specified, new runs will keep getting scheduled until this Schedule is paused or deleted. Already scheduled runs will be allowed to complete. Unset if not specified.
int64 max_run_count = 16
Optional. Maximum run count of the schedule. If specified, The schedule will be completed when either started_run_count >= max_run_count or when end_time is reached. If not specified, new runs will keep getting scheduled until this Schedule is paused or deleted. Already scheduled runs will be allowed to complete. Unset if not specified.
int64 started_run_count = 17
Output only. The number of runs started by this schedule.
Schedule.State state = 5
Output only. The state of this Schedule.
optional protobuf.Timestamp create_time = 6
Output only. Timestamp when this Schedule was created.
optional protobuf.Timestamp update_time = 19
Output only. Timestamp when this Schedule was updated.
optional protobuf.Timestamp next_run_time = 7
Output only. Timestamp when this Schedule should schedule the next run. Having a next_run_time in the past means the runs are being started behind schedule.
optional protobuf.Timestamp last_pause_time = 8
Output only. Timestamp when this Schedule was last paused. Unset if never paused.
optional protobuf.Timestamp last_resume_time = 9
Output only. Timestamp when this Schedule was last resumed. Unset if never resumed from pause.
int64 max_concurrent_run_count = 11
Required. Maximum number of runs that can be started concurrently for this Schedule. This is the limit for starting the scheduled requests and not the execution of the operations/jobs created by the requests (if applicable).
bool allow_queueing = 12
Optional. Whether new scheduled runs can be queued when max_concurrent_runs limit is reached. If set to true, new runs will be queued instead of skipped. Default to false.
bool catch_up = 13
Output only. Whether to backfill missed runs when the schedule is resumed from PAUSED state. If set to true, all missed runs will be scheduled. New runs will be scheduled after the backfill is complete. Default to false.
optional Schedule.RunResponse last_scheduled_run_response = 18
Output only. Response of the last scheduled run. This is the response for starting the scheduled requests and not the execution of the operations/jobs created by the requests (if applicable). Unset if no run has been scheduled yet.

Status of a scheduled run.

Used in: Schedule

optional protobuf.Timestamp scheduled_run_time = 1
The scheduled run time based on the user-specified schedule.
string run_response = 2
The response of the scheduled run.

Possible state of the schedule.

Used in: Schedule

STATE_UNSPECIFIED = 0
Unspecified.
ACTIVE = 1
The Schedule is active. Runs are being scheduled on the user-specified timespec.
PAUSED = 2
The schedule is paused. No new runs will be created until the schedule is resumed. Already started runs will be allowed to complete.
COMPLETED = 3
The Schedule is completed. No new runs will be scheduled. Already started runs will be allowed to complete. Schedules in completed state cannot be paused or resumed.

All parameters related to queuing and scheduling of custom jobs.

Used in: CustomJobSpec

optional protobuf.Duration timeout = 1
Optional. The maximum job running time. The default is 7 days.
bool restart_job_on_worker_restart = 3
Optional. Restarts the entire CustomJob if a worker gets restarted. This feature can be used by distributed training jobs that are not resilient to workers leaving and joining a job.
Scheduling.Strategy strategy = 4
Optional. This determines which type of scheduling strategy to use.
bool disable_retries = 5
Optional. Indicates if the job should retry for internal errors after the job starts running. If true, overrides `Scheduling.restart_job_on_worker_restart` to false.
optional protobuf.Duration max_wait_duration = 6
Optional. This is the maximum duration that a job will wait for the requested resources to be provisioned if the scheduling strategy is set to [Strategy.DWS_FLEX_START]. If set to 0, the job will wait indefinitely. The default is 24 hours.

Optional. This determines which type of scheduling strategy to use. Right now users have two options such as STANDARD which will use regular on demand resources to schedule the job, the other is SPOT which would leverage spot resources alongwith regular resources to schedule the job.

Used in: Scheduling

STRATEGY_UNSPECIFIED = 0
Strategy will default to STANDARD.
ON_DEMAND = 1
Deprecated. Regular on-demand provisioning strategy.
LOW_COST = 2
Deprecated. Low cost by making potential use of spot resources.
STANDARD = 3
Standard provisioning strategy uses regular on-demand resources.
SPOT = 4
Spot provisioning strategy uses spot resources.
FLEX_START = 6
Flex Start strategy uses DWS to queue for resources.

Schema is used to define the format of input/output data. Represents a select subset of an [OpenAPI 3.0 schema object](https://spec.openapis.org/oas/v3.0.3#schema-object). More fields may be added in the future as needed.

Used in: FunctionDeclaration, GenerationConfig

Type type = 1
Optional. The type of the data.
string format = 7
Optional. The format of the data. Supported formats: for NUMBER type: "float", "double" for INTEGER type: "int32", "int64" for STRING type: "email", "byte", etc
string title = 24
Optional. The title of the Schema.
string description = 8
Optional. The description of the data.
bool nullable = 6
Optional. Indicates if the value may be null.
optional protobuf.Value default = 23
Optional. Default value of the data.
optional Schema items = 2
Optional. SCHEMA FIELDS FOR TYPE ARRAY Schema of the elements of Type.ARRAY.
int64 min_items = 21
Optional. Minimum number of the elements for Type.ARRAY.
int64 max_items = 22
Optional. Maximum number of the elements for Type.ARRAY.
repeated string enum = 9
Optional. Possible values of the element of primitive type with enum format. Examples: 1. We can define direction as : {type:STRING, format:enum, enum:["EAST", NORTH", "SOUTH", "WEST"]} 2. We can define apartment number as : {type:INTEGER, format:enum, enum:["101", "201", "301"]}
map<string, Schema> properties = 3
Optional. SCHEMA FIELDS FOR TYPE OBJECT Properties of Type.OBJECT.
repeated string property_ordering = 25
Optional. The order of the properties. Not a standard field in open api spec. Only used to support the order of the properties.
repeated string required = 5
Optional. Required properties of Type.OBJECT.
int64 min_properties = 14
Optional. Minimum number of the properties for Type.OBJECT.
int64 max_properties = 15
Optional. Maximum number of the properties for Type.OBJECT.
double minimum = 16
Optional. SCHEMA FIELDS FOR TYPE INTEGER and NUMBER Minimum value of the Type.INTEGER and Type.NUMBER
double maximum = 17
Optional. Maximum value of the Type.INTEGER and Type.NUMBER
int64 min_length = 18
Optional. SCHEMA FIELDS FOR TYPE STRING Minimum length of the Type.STRING
int64 max_length = 19
Optional. Maximum length of the Type.STRING
string pattern = 20
Optional. Pattern of the Type.STRING to restrict a string to a regular expression.
optional protobuf.Value example = 4
Optional. Example of the object. Will only populated when the object is the root.
repeated Schema any_of = 11
Optional. The value should be validated against any (one or more) of the subschemas in the list.
string ref = 27
Optional. Allows indirect references between schema nodes. The value should be a valid reference to a child of the root `defs`. For example, the following schema defines a reference to a schema node named "Pet": type: object properties: pet: ref: #/defs/Pet defs: Pet: type: object properties: name: type: string The value of the "pet" property is a reference to the schema node named "Pet". See details in https://json-schema.org/understanding-json-schema/structuring
map<string, Schema> defs = 28
Optional. A map of definitions for use by `ref` Only allowed at the root of the schema.

Expression that allows ranking results based on annotation's property.

Used in: SearchDataItemsRequest

string saved_query = 1
Required. Saved query of the Annotation. Only Annotations belong to this saved query will be considered for ordering.
string order_by = 2
A comma-separated list of annotation fields to order by, sorted in ascending order. Use "desc" after a field name for descending. Must also specify saved_query.

Google search entry point.

Used in: GroundingMetadata

string rendered_content = 1
Optional. Web content snippet that can be embedded in a web page or an app webview.
bytes sdk_blob = 2
Optional. Base64 encoded JSON representing array of <search term, search url> tuple.

Stats requested for specific objective.

Used in: SearchModelDeploymentMonitoringStatsAnomaliesRequest

ModelDeploymentMonitoringObjectiveType type = 1
int32 top_feature_count = 4
If set, all attribution scores between [SearchModelDeploymentMonitoringStatsAnomaliesRequest.start_time][google.cloud.aiplatform.v1.SearchModelDeploymentMonitoringStatsAnomaliesRequest.start_time] and [SearchModelDeploymentMonitoringStatsAnomaliesRequest.end_time][google.cloud.aiplatform.v1.SearchModelDeploymentMonitoringStatsAnomaliesRequest.end_time] are fetched, and page token doesn't take effect in this case. Only used to retrieve attribution score for the top Features which has the highest attribution score in the latest monitoring run.

Represents an environment variable where the value is a secret in Cloud Secret Manager.

Used in: ReasoningEngineSpec.DeploymentSpec

string name = 1
Required. Name of the secret environment variable.
optional SecretRef secret_ref = 2
Required. Reference to a secret stored in the Cloud Secret Manager that will provide the value for this environment variable.

Reference to a secret stored in the Cloud Secret Manager that will provide the value for this environment variable.

Used in: SecretEnvVar

string secret = 1
Required. The name of the secret in Cloud Secret Manager. Format: {secret_name}.
string version = 2
The Cloud Secret Manager secret version. Can be 'latest' for the latest version, an integer for a specific version, or a version alias.

Segment of the content.

Used in: GroundingSupport

int32 part_index = 1
Output only. The index of a Part object within its parent Content object.
int32 start_index = 2
Output only. Start index in the given Part, measured in bytes. Offset from the start of the Part, inclusive, starting at zero.
int32 end_index = 3
Output only. End index in the given Part, measured in bytes. Offset from the start of the Part, exclusive, starting at zero.
string text = 4
Output only. The text corresponding to the segment from the response.

Configuration for the use of custom service account to run the workloads.

Used in: ResourceRuntimeSpec

bool enable_custom_service_account = 1
Required. If true, custom user-managed service account is enforced to run any workloads (for example, Vertex Jobs) on the resource. Otherwise, uses the [Vertex AI Custom Code Service Agent](https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents).
string service_account = 2
Optional. Required when all below conditions are met * `enable_custom_service_account` is true; * any runtime is specified via `ResourceRuntimeSpec` on creation time, for example, Ray The users must have `iam.serviceAccounts.actAs` permission on this service account and then the specified runtime containers will run as it. Do not set this field if you want to submit jobs using custom service account to this PersistentResource after creation, but only specify the `service_account` inside the job.

The SharePointSources to pass to ImportRagFiles.

repeated SharePointSources.SharePointSource share_point_sources = 1
The SharePoint sources.

An individual SharePointSource.

Used in: SharePointSources

oneof folder_source
The SharePoint folder source. If not provided, uses "root".
- string sharepoint_folder_path = 5
  The path of the SharePoint folder to download from.
- string sharepoint_folder_id = 6
  The ID of the SharePoint folder to download from.
oneof drive_source
The SharePoint drive source.
- string drive_name = 7
  The name of the drive to download from.
- string drive_id = 8
  The ID of the drive to download from.
string client_id = 1
The Application ID for the app registered in Microsoft Azure Portal. The application must also be configured with MS Graph permissions "Files.ReadAll", "Sites.ReadAll" and BrowserSiteLists.Read.All.
optional ApiAuth.ApiKeyConfig client_secret = 2
The application secret for the app registered in Azure.
string tenant_id = 3
Unique identifier of the Azure Active Directory Instance.
string sharepoint_site_name = 4
The name of the SharePoint site to download from. This can be the site name or the site id.
string file_id = 9
Output only. The SharePoint file id. Output only.

A set of Shielded Instance options. See [Images using supported Shielded VM features](https://cloud.google.com/compute/docs/instances/modifying-shielded-vm).

bool enable_secure_boot = 1
Defines whether the instance has [Secure Boot](https://cloud.google.com/compute/shielded-vm/docs/shielded-vm#secure-boot) enabled. Secure Boot helps ensure that the system only runs authentic software by verifying the digital signature of all boot components, and halting the boot process if signature verification fails.

The Slack source for the ImportRagFilesRequest.

repeated SlackSource.SlackChannels channels = 1
Required. The Slack channels.

SlackChannels contains the Slack channels and corresponding access token.

Used in: SlackSource

repeated SlackChannels.SlackChannel channels = 1
Required. The Slack channel IDs.
optional ApiAuth.ApiKeyConfig api_key_config = 3
Required. The SecretManager secret version resource name (e.g. projects/{project}/secrets/{secret}/versions/{version}) storing the Slack channel access token that has access to the slack channel IDs. See: https://api.slack.com/tutorials/tracks/getting-a-token.

SlackChannel contains the Slack channel ID and the time range to import.

Used in: SlackChannels

string channel_id = 1
Required. The Slack channel ID.
optional protobuf.Timestamp start_time = 2
Optional. The starting timestamp for messages to import.
optional protobuf.Timestamp end_time = 3
Optional. The ending timestamp for messages to import.

Config for SmoothGrad approximation of gradients. When enabled, the gradients are approximated by averaging the gradients from noisy samples in the vicinity of the inputs. Adding noise can help improve the computed gradients. Refer to this paper for more details: https://arxiv.org/pdf/1706.03825.pdf

Used in: IntegratedGradientsAttribution, XraiAttribution

oneof GradientNoiseSigma
Represents the standard deviation of the gaussian kernel that will be used to add noise to the interpolated inputs prior to computing gradients.
- float noise_sigma = 1
  This is a single float value and will be used to add noise to all the features. Use this field when all features are normalized to have the same distribution: scale to range [0, 1], [-1, 1] or z-scoring, where features are normalized to have 0-mean and 1-variance. Learn more about [normalization](https://developers.google.com/machine-learning/data-prep/transform/normalization). For best results the recommended value is about 10% - 20% of the standard deviation of the input feature. Refer to section 3.2 of the SmoothGrad paper: https://arxiv.org/pdf/1706.03825.pdf. Defaults to 0.1. If the distribution is different per feature, set [feature_noise_sigma][google.cloud.aiplatform.v1.SmoothGradConfig.feature_noise_sigma] instead for each feature.
- FeatureNoiseSigma feature_noise_sigma = 2
  This is similar to [noise_sigma][google.cloud.aiplatform.v1.SmoothGradConfig.noise_sigma], but provides additional flexibility. A separate noise sigma can be provided for each feature, which is useful if their distributions are different. No noise is added to features that are not set. If this field is unset, [noise_sigma][google.cloud.aiplatform.v1.SmoothGradConfig.noise_sigma] will be used for all features.
int32 noisy_sample_count = 3
The number of gradient samples to use for approximation. The higher this number, the more accurate the gradient is, but the runtime complexity increases by this factor as well. Valid range of its value is [1, 50]. Defaults to 3.

SpecialistPool represents customers' own workforce to work on their data labeling jobs. It includes a group of specialist managers and workers. Managers are responsible for managing the workers in this pool as well as customers' data labeling jobs associated with this pool. Customers create specialist pool as well as start data labeling jobs on Cloud, managers and workers handle the jobs using CrowdCompute console.

Used as response type in: SpecialistPoolService.GetSpecialistPool

Used as field type in: CreateSpecialistPoolRequest, ListSpecialistPoolsResponse, UpdateSpecialistPoolRequest

string name = 1
Required. The resource name of the SpecialistPool.
string display_name = 2
Required. The user-defined name of the SpecialistPool. The name can be up to 128 characters long and can consist of any UTF-8 characters. This field should be unique on project-level.
int32 specialist_managers_count = 3
Output only. The number of managers in this SpecialistPool.
repeated string specialist_manager_emails = 4
The email addresses of the managers in the SpecialistPool.
repeated string pending_data_labeling_jobs = 5
Output only. The resource name of the pending data labeling jobs.
repeated string specialist_worker_emails = 7
The email addresses of workers in the SpecialistPool.

Configuration for Speculative Decoding.

Used in: DeployedModel

oneof speculation
The type of speculation method to use.
- SpeculativeDecodingSpec.DraftModelSpeculation draft_model_speculation = 2
  draft model speculation.
- SpeculativeDecodingSpec.NgramSpeculation ngram_speculation = 3
  N-Gram speculation.
int32 speculative_token_count = 1
The number of speculative tokens to generate at each step.

Draft model speculation works by using the smaller model to generate candidate tokens for speculative decoding.

Used in: SpeculativeDecodingSpec

string draft_model = 1
Required. The resource name of the draft model.

N-Gram speculation works by trying to find matching tokens in the previous prompt sequence and use those as speculation for generating new tokens.

Used in: SpeculativeDecodingSpec

int32 ngram_size = 1
The number of last N input tokens used as ngram to search/match against the previous prompt sequence. This is equal to the N in N-Gram. The default value is 3 if not specified.

Metadata information for [NotebookService.StartNotebookRuntime][google.cloud.aiplatform.v1.NotebookService.StartNotebookRuntime].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.
string progress_message = 2
A human-readable message that shows the intermediate progress details of NotebookRuntime.

Response message for [NotebookService.StartNotebookRuntime][google.cloud.aiplatform.v1.NotebookService.StartNotebookRuntime].

(message has no fields)

Metadata information for [NotebookService.StopNotebookRuntime][google.cloud.aiplatform.v1.NotebookService.StopNotebookRuntime].

optional GenericOperationMetadata generic_metadata = 1
The operation generic information.

Response message for [NotebookService.StopNotebookRuntime][google.cloud.aiplatform.v1.NotebookService.StopNotebookRuntime].

(message has no fields)

Assigns input data to the training, validation, and test sets so that the distribution of values found in the categorical column (as specified by the `key` field) is mirrored within each split. The fraction values determine the relative sizes of the splits. For example, if the specified column has three values, with 50% of the rows having value "A", 25% value "B", and 25% value "C", and the split fractions are specified as 80/10/10, then the training set will constitute 80% of the training data, with about 50% of the training set rows having the value "A" for the specified column, about 25% having the value "B", and about 25% having the value "C". Only the top 500 occurring values are used; any values not in the top 500 values are randomly assigned to a split. If less than three rows contain a specific value, those rows are randomly assigned. Supported only for tabular Datasets.

Used in: InputDataConfig

double training_fraction = 1
The fraction of the input data that is to be used to train the Model.
double validation_fraction = 2
The fraction of the input data that is to be used to validate the Model.
double test_fraction = 3
The fraction of the input data that is to be used to evaluate the Model.
string key = 4
Required. The key is a name of one of the Dataset's data columns. The key provided must be for a categorical column.

Request message for [PredictionService.StreamingPredict][google.cloud.aiplatform.v1.PredictionService.StreamingPredict]. The first message must contain [endpoint][google.cloud.aiplatform.v1.StreamingPredictRequest.endpoint] field and optionally [input][]. The subsequent messages must contain [input][].

Used as request type in: PredictionService.ServerStreamingPredict, PredictionService.StreamingPredict

string endpoint = 1
Required. The name of the Endpoint requested to serve the prediction. Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
repeated Tensor inputs = 2
The prediction input.
optional Tensor parameters = 3
The parameters that govern the prediction.

Response message for [PredictionService.StreamingPredict][google.cloud.aiplatform.v1.PredictionService.StreamingPredict].

Used as response type in: PredictionService.ServerStreamingPredict, PredictionService.StreamingPredict

repeated Tensor outputs = 1
The prediction output.
optional Tensor parameters = 2
The parameters that govern the prediction.

A list of string values.

Used in: FeatureValue

repeated string values = 1
A list of string values.

One field of a Struct (or object) type feature value.

Used in: StructValue

string name = 1
Name of the field in the struct feature.
optional FeatureValue value = 2
The value for this field.

Struct (or object) type feature value.

Used in: FeatureValue

repeated StructFieldValue values = 1
A list of field values.

A message representing a Study.

Used as response type in: VizierService.CreateStudy, VizierService.GetStudy, VizierService.LookupStudy

Used as field type in: CreateStudyRequest, ListStudiesResponse

string name = 1
Output only. The name of a study. The study's globally unique identifier. Format: `projects/{project}/locations/{location}/studies/{study}`
string display_name = 2
Required. Describes the Study, default value is empty string.
optional StudySpec study_spec = 3
Required. Configuration of the Study.
Study.State state = 4
Output only. The detailed state of a Study.
optional protobuf.Timestamp create_time = 5
Output only. Time at which the study was created.
string inactive_reason = 6
Output only. A human readable reason why the Study is inactive. This should be empty if a study is ACTIVE or COMPLETED.

Describes the Study state.

Used in: Study, SuggestTrialsResponse

STATE_UNSPECIFIED = 0
The study state is unspecified.
ACTIVE = 1
The study is active.
INACTIVE = 2
The study is stopped due to an internal error.
COMPLETED = 3
The study is done when the service exhausts the parameter search space or max_trial_count is reached.

Represents specification of a Study.

Used in: HyperparameterTuningJob, Study

oneof automated_stopping_spec
- StudySpec.DecayCurveAutomatedStoppingSpec decay_curve_stopping_spec = 4
  The automated early stopping spec using decay curve rule.
- StudySpec.MedianAutomatedStoppingSpec median_automated_stopping_spec = 5
  The automated early stopping spec using median rule.
- StudySpec.ConvexAutomatedStoppingSpec convex_automated_stopping_spec = 9
  The automated early stopping spec using convex stopping rule.
repeated StudySpec.MetricSpec metrics = 1
Required. Metric specs for the Study.
repeated StudySpec.ParameterSpec parameters = 2
Required. The set of parameters to tune.
StudySpec.Algorithm algorithm = 3
The search algorithm specified for the Study.
StudySpec.ObservationNoise observation_noise = 6
The observation noise level of the study. Currently only supported by the Vertex AI Vizier service. Not supported by HyperparameterTuningJob or TrainingPipeline.
StudySpec.MeasurementSelectionType measurement_selection_type = 7
Describe which measurement selection type will be used
optional StudySpec.StudyStoppingConfig study_stopping_config = 11
Conditions for automated stopping of a Study. Enable automated stopping by configuring at least one condition.

The available search algorithms for the Study.

Used in: StudySpec

ALGORITHM_UNSPECIFIED = 0
The default algorithm used by Vertex AI for [hyperparameter tuning](https://cloud.google.com/vertex-ai/docs/training/hyperparameter-tuning-overview) and [Vertex AI Vizier](https://cloud.google.com/vertex-ai/docs/vizier).
GRID_SEARCH = 2
Simple grid search within the feasible space. To use grid search, all parameters must be `INTEGER`, `CATEGORICAL`, or `DISCRETE`.
RANDOM_SEARCH = 3
Simple random search within the feasible space.

Configuration for ConvexAutomatedStoppingSpec. When there are enough completed trials (configured by min_measurement_count), for pending trials with enough measurements and steps, the policy first computes an overestimate of the objective value at max_num_steps according to the slope of the incomplete objective value curve. No prediction can be made if the curve is completely flat. If the overestimation is worse than the best objective value of the completed trials, this pending trial will be early-stopped, but a last measurement will be added to the pending trial with max_num_steps and predicted objective value from the autoregression model.

Used in: StudySpec

int64 max_step_count = 1
Steps used in predicting the final objective for early stopped trials. In general, it's set to be the same as the defined steps in training / tuning. If not defined, it will learn it from the completed trials. When use_steps is false, this field is set to the maximum elapsed seconds.
int64 min_step_count = 2
Minimum number of steps for a trial to complete. Trials which do not have a measurement with step_count > min_step_count won't be considered for early stopping. It's ok to set it to 0, and a trial can be early stopped at any stage. By default, min_step_count is set to be one-tenth of the max_step_count. When use_elapsed_duration is true, this field is set to the minimum elapsed seconds.
int64 min_measurement_count = 3
The minimal number of measurements in a Trial. Early-stopping checks will not trigger if less than min_measurement_count+1 completed trials or pending trials with less than min_measurement_count measurements. If not defined, the default value is 5.
string learning_rate_parameter_name = 4
The hyper-parameter name used in the tuning job that stands for learning rate. Leave it blank if learning rate is not in a parameter in tuning. The learning_rate is used to estimate the objective value of the ongoing trial.
bool use_elapsed_duration = 5
This bool determines whether or not the rule is applied based on elapsed_secs or steps. If use_elapsed_duration==false, the early stopping decision is made according to the predicted objective values according to the target steps. If use_elapsed_duration==true, elapsed_secs is used instead of steps. Also, in this case, the parameters max_num_steps and min_num_steps are overloaded to contain max_elapsed_seconds and min_elapsed_seconds.
optional bool update_all_stopped_trials = 6
ConvexAutomatedStoppingSpec by default only updates the trials that needs to be early stopped using a newly trained auto-regressive model. When this flag is set to True, all stopped trials from the beginning are potentially updated in terms of their `final_measurement`. Also, note that the training logic of autoregressive models is different in this case. Enabling this option has shown better results and this may be the default option in the future.

The decay curve automated stopping rule builds a Gaussian Process Regressor to predict the final objective value of a Trial based on the already completed Trials and the intermediate measurements of the current Trial. Early stopping is requested for the current Trial if there is very low probability to exceed the optimal value found so far.

Used in: StudySpec

bool use_elapsed_duration = 1
True if [Measurement.elapsed_duration][google.cloud.aiplatform.v1.Measurement.elapsed_duration] is used as the x-axis of each Trials Decay Curve. Otherwise, [Measurement.step_count][google.cloud.aiplatform.v1.Measurement.step_count] will be used as the x-axis.

This indicates which measurement to use if/when the service automatically selects the final measurement from previously reported intermediate measurements. Choose this based on two considerations: A) Do you expect your measurements to monotonically improve? If so, choose LAST_MEASUREMENT. On the other hand, if you're in a situation where your system can "over-train" and you expect the performance to get better for a while but then start declining, choose BEST_MEASUREMENT. B) Are your measurements significantly noisy and/or irreproducible? If so, BEST_MEASUREMENT will tend to be over-optimistic, and it may be better to choose LAST_MEASUREMENT. If both or neither of (A) and (B) apply, it doesn't matter which selection type is chosen.

Used in: StudySpec

MEASUREMENT_SELECTION_TYPE_UNSPECIFIED = 0
Will be treated as LAST_MEASUREMENT.
LAST_MEASUREMENT = 1
Use the last measurement reported.
BEST_MEASUREMENT = 2
Use the best measurement reported.

The median automated stopping rule stops a pending Trial if the Trial's best objective_value is strictly below the median 'performance' of all completed Trials reported up to the Trial's last measurement. Currently, 'performance' refers to the running average of the objective values reported by the Trial in each measurement.

Used in: StudySpec

bool use_elapsed_duration = 1
True if median automated stopping rule applies on [Measurement.elapsed_duration][google.cloud.aiplatform.v1.Measurement.elapsed_duration]. It means that elapsed_duration field of latest measurement of current Trial is used to compute median objective value for each completed Trials.

Represents a metric to optimize.

Used in: StudySpec

string metric_id = 1
Required. The ID of the metric. Must not contain whitespaces and must be unique amongst all MetricSpecs.
MetricSpec.GoalType goal = 2
Required. The optimization goal of the metric.
optional MetricSpec.SafetyMetricConfig safety_config = 3
Used for safe search. In the case, the metric will be a safety metric. You must provide a separate metric for objective metric.

The available types of optimization goals.

Used in: MetricSpec

GOAL_TYPE_UNSPECIFIED = 0
Goal Type will default to maximize.
MAXIMIZE = 1
Maximize the goal metric.
MINIMIZE = 2
Minimize the goal metric.

Used in safe optimization to specify threshold levels and risk tolerance.

Used in: MetricSpec

double safety_threshold = 1
Safety threshold (boundary value between safe and unsafe). NOTE that if you leave SafetyMetricConfig unset, a default value of 0 will be used.
optional double desired_min_safe_trials_fraction = 2
Desired minimum fraction of safe trials (over total number of trials) that should be targeted by the algorithm at any time during the study (best effort). This should be between 0.0 and 1.0 and a value of 0.0 means that there is no minimum and an algorithm proceeds without targeting any specific fraction. A value of 1.0 means that the algorithm attempts to only Suggest safe Trials.

Describes the noise level of the repeated observations. "Noisy" means that the repeated observations with the same Trial parameters may lead to different metric evaluations.

Used in: StudySpec

OBSERVATION_NOISE_UNSPECIFIED = 0
The default noise level chosen by Vertex AI.
LOW = 1
Vertex AI assumes that the objective function is (nearly) perfectly reproducible, and will never repeat the same Trial parameters.
HIGH = 2
Vertex AI will estimate the amount of noise in metric evaluations, it may repeat the same Trial parameters more than once.

Represents a single parameter to optimize.

Used in: StudySpec, ParameterSpec.ConditionalParameterSpec

oneof parameter_value_spec
- ParameterSpec.DoubleValueSpec double_value_spec = 2
  The value spec for a 'DOUBLE' parameter.
- ParameterSpec.IntegerValueSpec integer_value_spec = 3
  The value spec for an 'INTEGER' parameter.
- ParameterSpec.CategoricalValueSpec categorical_value_spec = 4
  The value spec for a 'CATEGORICAL' parameter.
- ParameterSpec.DiscreteValueSpec discrete_value_spec = 5
  The value spec for a 'DISCRETE' parameter.
string parameter_id = 1
Required. The ID of the parameter. Must not contain whitespaces and must be unique amongst all ParameterSpecs.
ParameterSpec.ScaleType scale_type = 6
How the parameter should be scaled. Leave unset for `CATEGORICAL` parameters.
repeated ParameterSpec.ConditionalParameterSpec conditional_parameter_specs = 10
A conditional parameter node is active if the parameter's value matches the conditional node's parent_value_condition. If two items in conditional_parameter_specs have the same name, they must have disjoint parent_value_condition.

Value specification for a parameter in `CATEGORICAL` type.

Used in: ParameterSpec

repeated string values = 1
Required. The list of possible categories.
optional string default_value = 3
A default value for a `CATEGORICAL` parameter that is assumed to be a relatively good starting point. Unset value signals that there is no offered starting point. Currently only supported by the Vertex AI Vizier service. Not supported by HyperparameterTuningJob or TrainingPipeline.

Represents a parameter spec with condition from its parent parameter.

Used in: ParameterSpec

oneof parent_value_condition
A set of parameter values from the parent ParameterSpec's feasible space.
- ConditionalParameterSpec.DiscreteValueCondition parent_discrete_values = 2
  The spec for matching values from a parent parameter of `DISCRETE` type.
- ConditionalParameterSpec.IntValueCondition parent_int_values = 3
  The spec for matching values from a parent parameter of `INTEGER` type.
- ConditionalParameterSpec.CategoricalValueCondition parent_categorical_values = 4
  The spec for matching values from a parent parameter of `CATEGORICAL` type.
optional ParameterSpec parameter_spec = 1
Required. The spec for a conditional parameter.

Represents the spec to match categorical values from parent parameter.

Used in: ConditionalParameterSpec

repeated string values = 1
Required. Matches values of the parent parameter of 'CATEGORICAL' type. All values must exist in `categorical_value_spec` of parent parameter.

Represents the spec to match discrete values from parent parameter.

Used in: ConditionalParameterSpec

repeated double values = 1
Required. Matches values of the parent parameter of 'DISCRETE' type. All values must exist in `discrete_value_spec` of parent parameter. The Epsilon of the value matching is 1e-10.

Represents the spec to match integer values from parent parameter.

Used in: ConditionalParameterSpec

repeated int64 values = 1
Required. Matches values of the parent parameter of 'INTEGER' type. All values must lie in `integer_value_spec` of parent parameter.

Value specification for a parameter in `DISCRETE` type.

Used in: ParameterSpec

repeated double values = 1
Required. A list of possible values. The list should be in increasing order and at least 1e-10 apart. 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.
optional double default_value = 3
A default value for a `DISCRETE` parameter that is assumed to be a relatively good starting point. Unset value signals that there is no offered starting point. It automatically rounds to the nearest feasible discrete point. Currently only supported by the Vertex AI Vizier service. Not supported by HyperparameterTuningJob or TrainingPipeline.

Value specification for a parameter in `DOUBLE` type.

Used in: ParameterSpec

double min_value = 1
Required. Inclusive minimum value of the parameter.
double max_value = 2
Required. Inclusive maximum value of the parameter.
optional double default_value = 4
A default value for a `DOUBLE` parameter that is assumed to be a relatively good starting point. Unset value signals that there is no offered starting point. Currently only supported by the Vertex AI Vizier service. Not supported by HyperparameterTuningJob or TrainingPipeline.

Value specification for a parameter in `INTEGER` type.

Used in: ParameterSpec

int64 min_value = 1
Required. Inclusive minimum value of the parameter.
int64 max_value = 2
Required. Inclusive maximum value of the parameter.
optional int64 default_value = 4
A default value for an `INTEGER` parameter that is assumed to be a relatively good starting point. Unset value signals that there is no offered starting point. Currently only supported by the Vertex AI Vizier service. Not supported by HyperparameterTuningJob or TrainingPipeline.

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

Used in: ParameterSpec

SCALE_TYPE_UNSPECIFIED = 0
By default, no scaling is applied.
UNIT_LINEAR_SCALE = 1
Scales the feasible space to (0, 1) linearly.
UNIT_LOG_SCALE = 2
Scales the feasible space logarithmically to (0, 1). The entire feasible space must be strictly positive.
UNIT_REVERSE_LOG_SCALE = 3
Scales 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.

The configuration (stopping conditions) for automated stopping of a Study. Conditions include trial budgets, time budgets, and convergence detection.

Used in: StudySpec

optional protobuf.BoolValue should_stop_asap = 1
If true, a Study enters STOPPING_ASAP whenever it would normally enters STOPPING state. The bottom line is: set to true if you want to interrupt on-going evaluations of Trials as soon as the study stopping condition is met. (Please see Study.State documentation for the source of truth).
optional StudyTimeConstraint minimum_runtime_constraint = 2
Each "stopping rule" in this proto specifies an "if" condition. Before Vizier would generate a new suggestion, it first checks each specified stopping rule, from top to bottom in this list. Note that the first few rules (e.g. minimum_runtime_constraint, min_num_trials) will prevent other stopping rules from being evaluated until they are met. For example, setting `min_num_trials=5` and `always_stop_after= 1 hour` means that the Study will ONLY stop after it has 5 COMPLETED trials, even if more than an hour has passed since its creation. It follows the first applicable rule (whose "if" condition is satisfied) to make a stopping decision. If none of the specified rules are applicable, then Vizier decides that the study should not stop. If Vizier decides that the study should stop, the study enters STOPPING state (or STOPPING_ASAP if should_stop_asap = true). IMPORTANT: The automatic study state transition happens precisely as described above; that is, deleting trials or updating StudyConfig NEVER automatically moves the study state back to ACTIVE. If you want to _resume_ a Study that was stopped, 1) change the stopping conditions if necessary, 2) activate the study, and then 3) ask for suggestions. If the specified time or duration has not passed, do not stop the study.
optional StudyTimeConstraint maximum_runtime_constraint = 3
If the specified time or duration has passed, stop the study.
optional protobuf.Int32Value min_num_trials = 4
If there are fewer than this many COMPLETED trials, do not stop the study.
optional protobuf.Int32Value max_num_trials = 5
If there are more than this many trials, stop the study.
optional protobuf.Int32Value max_num_trials_no_progress = 6
If the objective value has not improved for this many consecutive trials, stop the study. WARNING: Effective only for single-objective studies.
optional protobuf.Duration max_duration_no_progress = 7
If the objective value has not improved for this much time, stop the study. WARNING: Effective only for single-objective studies.

Time-based Constraint for Study

Used in: StudySpec.StudyStoppingConfig

oneof constraint
- protobuf.Duration max_duration = 1
  Counts the wallclock time passed since the creation of this Study.
- protobuf.Timestamp end_time = 2
  Compares the wallclock time to this time. Must use UTC timezone.

Details of operations that perform Trials suggestion.

optional GenericOperationMetadata generic_metadata = 1
Operation metadata for suggesting Trials.
string client_id = 2
The identifier of the client that is requesting the suggestion. If multiple SuggestTrialsRequests have the same `client_id`, the service will return the identical suggested Trial if the Trial is pending, and provide a new Trial if the last suggested Trial was completed.

Response message for [VizierService.SuggestTrials][google.cloud.aiplatform.v1.VizierService.SuggestTrials].

repeated Trial trials = 1
A list of Trials.
Study.State study_state = 2
The state of the Study.
optional protobuf.Timestamp start_time = 3
The time at which the operation was started.
optional protobuf.Timestamp end_time = 4
The time at which operation processing completed.

Input for summarization helpfulness metric.

optional SummarizationHelpfulnessSpec metric_spec = 1
Required. Spec for summarization helpfulness score metric.
optional SummarizationHelpfulnessInstance instance = 2
Required. Summarization helpfulness instance.

Spec for summarization helpfulness instance.

Used in: SummarizationHelpfulnessInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Optional. Ground truth used to compare against the prediction.
optional string context = 3
Required. Text to be summarized.
optional string instruction = 4
Optional. Summarization prompt for LLM.

Spec for summarization helpfulness result.

optional float score = 1
Output only. Summarization Helpfulness score.
string explanation = 2
Output only. Explanation for summarization helpfulness score.
optional float confidence = 3
Output only. Confidence for summarization helpfulness score.

Spec for summarization helpfulness score metric.

Used in: SummarizationHelpfulnessInput

bool use_reference = 1
Optional. Whether to use instance.reference to compute summarization helpfulness.
int32 version = 2
Optional. Which version to use for evaluation.

Input for summarization quality metric.

optional SummarizationQualitySpec metric_spec = 1
Required. Spec for summarization quality score metric.
optional SummarizationQualityInstance instance = 2
Required. Summarization quality instance.

Spec for summarization quality instance.

Used in: SummarizationQualityInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Optional. Ground truth used to compare against the prediction.
optional string context = 3
Required. Text to be summarized.
optional string instruction = 4
Required. Summarization prompt for LLM.

Spec for summarization quality result.

optional float score = 1
Output only. Summarization Quality score.
string explanation = 2
Output only. Explanation for summarization quality score.
optional float confidence = 3
Output only. Confidence for summarization quality score.

Spec for summarization quality score metric.

Used in: SummarizationQualityInput

bool use_reference = 1
Optional. Whether to use instance.reference to compute summarization quality.
int32 version = 2
Optional. Which version to use for evaluation.

Input for summarization verbosity metric.

optional SummarizationVerbositySpec metric_spec = 1
Required. Spec for summarization verbosity score metric.
optional SummarizationVerbosityInstance instance = 2
Required. Summarization verbosity instance.

Spec for summarization verbosity instance.

Used in: SummarizationVerbosityInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Optional. Ground truth used to compare against the prediction.
optional string context = 3
Required. Text to be summarized.
optional string instruction = 4
Optional. Summarization prompt for LLM.

Spec for summarization verbosity result.

Used in: ModelMonitoringObjectiveConfig.PredictionDriftDetectionConfig, ModelMonitoringObjectiveConfig.TrainingPredictionSkewDetectionConfig, ModelMonitoringStatsAnomalies.FeatureHistoricStatsAnomalies

optional float score = 1
Output only. Summarization Verbosity score.
string explanation = 2
Output only. Explanation for summarization verbosity score.
optional float confidence = 3
Output only. Confidence for summarization verbosity score.

Spec for summarization verbosity score metric.

Used in: SummarizationVerbosityInput

bool use_reference = 1
Optional. Whether to use instance.reference to compute summarization verbosity.
int32 version = 2
Optional. Which version to use for evaluation.

Hyperparameters for SFT.

Used in: SupervisedTuningSpec

int64 epoch_count = 1
Optional. Number of complete passes the model makes over the entire training dataset during training.
double learning_rate_multiplier = 2
Optional. Multiplier for adjusting the default learning rate.
SupervisedHyperParameters.AdapterSize adapter_size = 3
Optional. Adapter size for tuning.

Supported adapter sizes for tuning.

Used in: SupervisedHyperParameters

ADAPTER_SIZE_UNSPECIFIED = 0
Adapter size is unspecified.
ADAPTER_SIZE_ONE = 1
Adapter size 1.
ADAPTER_SIZE_FOUR = 2
Adapter size 4.
ADAPTER_SIZE_EIGHT = 3
Adapter size 8.
ADAPTER_SIZE_SIXTEEN = 4
Adapter size 16.

Tuning data statistics for Supervised Tuning.

Used in: TuningDataStats

int64 tuning_dataset_example_count = 1
Output only. Number of examples in the tuning dataset.
int64 total_tuning_character_count = 2
Output only. Number of tuning characters in the tuning dataset.
int64 total_billable_character_count = 3
Output only. Number of billable characters in the tuning dataset.
int64 total_billable_token_count = 9
Output only. Number of billable tokens in the tuning dataset.
int64 tuning_step_count = 4
Output only. Number of tuning steps for this Tuning Job.
optional SupervisedTuningDatasetDistribution user_input_token_distribution = 5
Output only. Dataset distributions for the user input tokens.
optional SupervisedTuningDatasetDistribution user_output_token_distribution = 6
Output only. Dataset distributions for the user output tokens.
optional SupervisedTuningDatasetDistribution user_message_per_example_distribution = 7
Output only. Dataset distributions for the messages per example.
repeated Content user_dataset_examples = 8
Output only. Sample user messages in the training dataset uri.
int64 total_truncated_example_count = 10
The number of examples in the dataset that have been truncated by any amount.
repeated int64 truncated_example_indices = 11
A partial sample of the indices (starting from 1) of the truncated examples.

Dataset distribution for Supervised Tuning.

Used in: SupervisedTuningDataStats

int64 sum = 1
Output only. Sum of a given population of values.
int64 billable_sum = 9
Output only. Sum of a given population of values that are billable.
double min = 2
Output only. The minimum of the population values.
double max = 3
Output only. The maximum of the population values.
double mean = 4
Output only. The arithmetic mean of the values in the population.
double median = 5
Output only. The median of the values in the population.
double p5 = 6
Output only. The 5th percentile of the values in the population.
double p95 = 7
Output only. The 95th percentile of the values in the population.
repeated SupervisedTuningDatasetDistribution.DatasetBucket buckets = 8
Output only. Defines the histogram bucket.

Dataset bucket used to create a histogram for the distribution given a population of values.

Used in: SupervisedTuningDatasetDistribution

double count = 1
Output only. Number of values in the bucket.
double left = 2
Output only. Left bound of the bucket.
double right = 3
Output only. Right bound of the bucket.

Tuning Spec for Supervised Tuning for first party models.

Used in: TuningJob

string training_dataset_uri = 1
Required. Cloud Storage path to file containing training dataset for tuning. The dataset must be formatted as a JSONL file.
string validation_dataset_uri = 2
Optional. Cloud Storage path to file containing validation dataset for tuning. The dataset must be formatted as a JSONL file.
optional SupervisedHyperParameters hyper_parameters = 3
Optional. Hyperparameters for SFT.

The storage details for TFRecord output content.

Used in: FeatureValueDestination

optional GcsDestination gcs_destination = 1
Required. Google Cloud Storage location.

A tensor value type.

Used in: DirectPredictRequest, DirectPredictResponse, StreamDirectPredictRequest, StreamDirectPredictResponse, StreamingPredictRequest, StreamingPredictResponse

Tensor.DataType dtype = 1
The data type of tensor.
repeated int64 shape = 2
Shape of the tensor.
repeated bool bool_val = 3
Type specific representations that make it easy to create tensor protos in all languages. Only the representation corresponding to "dtype" can be set. The values hold the flattened representation of the tensor in row major order. [BOOL][google.cloud.aiplatform.v1.Tensor.DataType.BOOL]
repeated string string_val = 14
[STRING][google.cloud.aiplatform.v1.Tensor.DataType.STRING]
repeated bytes bytes_val = 15
[STRING][google.cloud.aiplatform.v1.Tensor.DataType.STRING]
repeated float float_val = 5
[FLOAT][google.cloud.aiplatform.v1.Tensor.DataType.FLOAT]
repeated double double_val = 6
[DOUBLE][google.cloud.aiplatform.v1.Tensor.DataType.DOUBLE]
repeated int32 int_val = 7
[INT_8][google.cloud.aiplatform.v1.Tensor.DataType.INT8] [INT_16][google.cloud.aiplatform.v1.Tensor.DataType.INT16] [INT_32][google.cloud.aiplatform.v1.Tensor.DataType.INT32]
repeated int64 int64_val = 8
[INT64][google.cloud.aiplatform.v1.Tensor.DataType.INT64]
repeated uint32 uint_val = 9
[UINT8][google.cloud.aiplatform.v1.Tensor.DataType.UINT8] [UINT16][google.cloud.aiplatform.v1.Tensor.DataType.UINT16] [UINT32][google.cloud.aiplatform.v1.Tensor.DataType.UINT32]
repeated uint64 uint64_val = 10
[UINT64][google.cloud.aiplatform.v1.Tensor.DataType.UINT64]
repeated Tensor list_val = 11
A list of tensor values.
map<string, Tensor> struct_val = 12
A map of string to tensor.
bytes tensor_val = 13
Serialized raw tensor content.

Data type of the tensor.

Used in: Tensor

DATA_TYPE_UNSPECIFIED = 0
Not a legal value for DataType. Used to indicate a DataType field has not been set.
BOOL = 1
Data types that all computation devices are expected to be capable to support.
STRING = 2
FLOAT = 3
DOUBLE = 4
INT8 = 5
INT16 = 6
INT32 = 7
INT64 = 8
UINT8 = 9
UINT16 = 10
UINT32 = 11
UINT64 = 12

Tensorboard is a physical database that stores users' training metrics. A default Tensorboard is provided in each region of a Google Cloud project. If needed users can also create extra Tensorboards in their projects.

Used as response type in: TensorboardService.GetTensorboard

Used as field type in: CreateTensorboardRequest, ListTensorboardsResponse, UpdateTensorboardRequest

string name = 1
Output only. Name of the Tensorboard. Format: `projects/{project}/locations/{location}/tensorboards/{tensorboard}`
string display_name = 2
Required. User provided name of this Tensorboard.
string description = 3
Description of this Tensorboard.
optional EncryptionSpec encryption_spec = 11
Customer-managed encryption key spec for a Tensorboard. If set, this Tensorboard and all sub-resources of this Tensorboard will be secured by this key.
string blob_storage_path_prefix = 10
Output only. Consumer project Cloud Storage path prefix used to store blob data, which can either be a bucket or directory. Does not end with a '/'.
int32 run_count = 5
Output only. The number of Runs stored in this Tensorboard.
optional protobuf.Timestamp create_time = 6
Output only. Timestamp when this Tensorboard was created.
optional protobuf.Timestamp update_time = 7
Output only. Timestamp when this Tensorboard was last updated.
map<string, string> labels = 8
The labels with user-defined metadata to organize your Tensorboards. 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. No more than 64 user labels can be associated with one Tensorboard (System labels are excluded). See https://goo.gl/xmQnxf for more information and examples of labels. System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
string etag = 9
Used to perform a consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
bool is_default = 12
Used to indicate if the TensorBoard instance is the default one. Each project & region can have at most one default TensorBoard instance. Creation of a default TensorBoard instance and updating an existing TensorBoard instance to be default will mark all other TensorBoard instances (if any) as non default.
bool satisfies_pzs = 13
Output only. Reserved for future use.
bool satisfies_pzi = 14
Output only. Reserved for future use.

One blob (e.g, image, graph) viewable on a blob metric plot.

Used in: ReadTensorboardBlobDataResponse, TensorboardBlobSequence

string id = 1
Output only. A URI safe key uniquely identifying a blob. Can be used to locate the blob stored in the Cloud Storage bucket of the consumer project.
bytes data = 2
Optional. The bytes of the blob is not present unless it's returned by the ReadTensorboardBlobData endpoint.

One point viewable on a blob metric plot, but mostly just a wrapper message to work around repeated fields can't be used directly within `oneof` fields.

Used in: TimeSeriesDataPoint

repeated TensorboardBlob values = 1
List of blobs contained within the sequence.

A TensorboardExperiment is a group of TensorboardRuns, that are typically the results of a training job run, in a Tensorboard.

Used as response type in: TensorboardService.CreateTensorboardExperiment, TensorboardService.GetTensorboardExperiment, TensorboardService.UpdateTensorboardExperiment

Used as field type in: CreateTensorboardExperimentRequest, ListTensorboardExperimentsResponse, UpdateTensorboardExperimentRequest

string name = 1
Output only. Name of the TensorboardExperiment. Format: `projects/{project}/locations/{location}/tensorboards/{tensorboard}/experiments/{experiment}`
string display_name = 2
User provided name of this TensorboardExperiment.
string description = 3
Description of this TensorboardExperiment.
optional protobuf.Timestamp create_time = 4
Output only. Timestamp when this TensorboardExperiment was created.
optional protobuf.Timestamp update_time = 5
Output only. Timestamp when this TensorboardExperiment was last updated.
map<string, string> labels = 6
The labels with user-defined metadata to organize your TensorboardExperiment. Label keys and values cannot be longer than 64 characters (Unicode codepoints), can only contain lowercase letters, numeric characters, underscores and dashes. International characters are allowed. No more than 64 user labels can be associated with one Dataset (System labels are excluded). See https://goo.gl/xmQnxf for more information and examples of labels. System reserved label keys are prefixed with `aiplatform.googleapis.com/` and are immutable. The following system labels exist for each Dataset: * `aiplatform.googleapis.com/dataset_metadata_schema`: output only. Its value is the [metadata_schema's][google.cloud.aiplatform.v1.Dataset.metadata_schema_uri] title.
string etag = 7
Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
string source = 8
Immutable. Source of the TensorboardExperiment. Example: a custom training job.

TensorboardRun maps to a specific execution of a training job with a given set of hyperparameter values, model definition, dataset, etc

Used as response type in: TensorboardService.CreateTensorboardRun, TensorboardService.GetTensorboardRun, TensorboardService.UpdateTensorboardRun

Used as field type in: BatchCreateTensorboardRunsResponse, CreateTensorboardRunRequest, ListTensorboardRunsResponse, UpdateTensorboardRunRequest

string name = 1
Output only. Name of the TensorboardRun. Format: `projects/{project}/locations/{location}/tensorboards/{tensorboard}/experiments/{experiment}/runs/{run}`
string display_name = 2
Required. User provided name of this TensorboardRun. This value must be unique among all TensorboardRuns belonging to the same parent TensorboardExperiment.
string description = 3
Description of this TensorboardRun.
optional protobuf.Timestamp create_time = 6
Output only. Timestamp when this TensorboardRun was created.
optional protobuf.Timestamp update_time = 7
Output only. Timestamp when this TensorboardRun was last updated.
map<string, string> labels = 8
The labels with user-defined metadata to organize your TensorboardRuns. This field will be used to filter and visualize Runs in the Tensorboard UI. For example, a Vertex AI training job can set a label aiplatform.googleapis.com/training_job_id=xxxxx to all the runs created within that job. An end user can set a label experiment_id=xxxxx for all the runs produced in a Jupyter notebook. These runs can be grouped by a label value and visualized together in the Tensorboard UI. 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. No more than 64 user labels can be associated with one TensorboardRun (System labels are excluded). See https://goo.gl/xmQnxf for more information and examples of labels. System reserved label keys are prefixed with "aiplatform.googleapis.com/" and are immutable.
string etag = 9
Used to perform a consistent read-modify-write updates. If not set, a blind "overwrite" update happens.

One point viewable on a tensor metric plot.

Used in: TimeSeriesDataPoint

bytes value = 1
Required. Serialized form of https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/framework/tensor.proto
int32 version_number = 2
Optional. Version number of TensorProto used to serialize [value][google.cloud.aiplatform.v1.TensorboardTensor.value].

TensorboardTimeSeries maps to times series produced in training runs

Used as response type in: TensorboardService.CreateTensorboardTimeSeries, TensorboardService.GetTensorboardTimeSeries, TensorboardService.UpdateTensorboardTimeSeries

Used as field type in: BatchCreateTensorboardTimeSeriesResponse, CreateTensorboardTimeSeriesRequest, ListTensorboardTimeSeriesResponse, UpdateTensorboardTimeSeriesRequest

string name = 1
Output only. Name of the TensorboardTimeSeries.
string display_name = 2
Required. User provided name of this TensorboardTimeSeries. This value should be unique among all TensorboardTimeSeries resources belonging to the same TensorboardRun resource (parent resource).
string description = 3
Description of this TensorboardTimeSeries.
TensorboardTimeSeries.ValueType value_type = 4
Required. Immutable. Type of TensorboardTimeSeries value.
optional protobuf.Timestamp create_time = 5
Output only. Timestamp when this TensorboardTimeSeries was created.
optional protobuf.Timestamp update_time = 6
Output only. Timestamp when this TensorboardTimeSeries was last updated.
string etag = 7
Used to perform a consistent read-modify-write updates. If not set, a blind "overwrite" update happens.
string plugin_name = 8
Immutable. Name of the plugin this time series pertain to. Such as Scalar, Tensor, Blob
bytes plugin_data = 9
Data of the current plugin, with the size limited to 65KB.
optional TensorboardTimeSeries.Metadata metadata = 10
Output only. Scalar, Tensor, or Blob metadata for this TensorboardTimeSeries.

Describes metadata for a TensorboardTimeSeries.

Used in: TensorboardTimeSeries

int64 max_step = 1
Output only. Max step index of all data points within a TensorboardTimeSeries.
optional protobuf.Timestamp max_wall_time = 2
Output only. Max wall clock timestamp of all data points within a TensorboardTimeSeries.
int64 max_blob_sequence_length = 3
Output only. The largest blob sequence length (number of blobs) of all data points in this time series, if its ValueType is BLOB_SEQUENCE.

An enum representing the value type of a TensorboardTimeSeries.

Used in: TensorboardTimeSeries, TimeSeriesData

VALUE_TYPE_UNSPECIFIED = 0
The value type is unspecified.
SCALAR = 1
Used for TensorboardTimeSeries that is a list of scalars. E.g. accuracy of a model over epochs/time.
TENSOR = 2
Used for TensorboardTimeSeries that is a list of tensors. E.g. histograms of weights of layer in a model over epoch/time.
BLOB_SEQUENCE = 3
Used for TensorboardTimeSeries that is a list of blob sequences. E.g. set of sample images with labels over epochs/time.

The config for feature monitoring threshold.

oneof threshold
- double value = 1
  Specify a threshold value that can trigger the alert. If this threshold config is for feature distribution distance: 1. For categorical feature, the distribution distance is calculated by L-inifinity norm. 2. For numerical feature, the distribution distance is calculated by Jensen–Shannon divergence. Each feature must have a non-zero threshold if they need to be monitored. Otherwise no alert will be triggered for that feature.

All the data stored in a TensorboardTimeSeries.

Used in: BatchReadTensorboardTimeSeriesDataResponse, ReadTensorboardTimeSeriesDataResponse, WriteTensorboardRunDataRequest

string tensorboard_time_series_id = 1
Required. The ID of the TensorboardTimeSeries, which will become the final component of the TensorboardTimeSeries' resource name
TensorboardTimeSeries.ValueType value_type = 2
Required. Immutable. The value type of this time series. All the values in this time series data must match this value type.
repeated TimeSeriesDataPoint values = 3
Required. Data points in this time series.

A TensorboardTimeSeries data point.

Used in: ExportTensorboardTimeSeriesDataResponse, TimeSeriesData

oneof value
Value of this time series data point.
- Scalar scalar = 3
  A scalar value.
- TensorboardTensor tensor = 4
  A tensor value.
- TensorboardBlobSequence blobs = 5
  A blob sequence value.
optional protobuf.Timestamp wall_time = 1
Wall clock timestamp when this data point is generated by the end user.
int64 step = 2
Step index of this data point within the run.

Assigns input data to training, validation, and test sets based on a provided timestamps. The youngest data pieces are assigned to training set, next to validation set, and the oldest to the test set. Supported only for tabular Datasets.

Used in: InputDataConfig

double training_fraction = 1
The fraction of the input data that is to be used to train the Model.
double validation_fraction = 2
The fraction of the input data that is to be used to validate the Model.
double test_fraction = 3
The fraction of the input data that is to be used to evaluate the Model.
string key = 4
Required. The key is a name of one of the Dataset's data columns. The values of the key (the values in the column) must be in RFC 3339 `date-time` format, where `time-offset` = `"Z"` (e.g. 1985-04-12T23:20:50.52Z). If for a piece of data the key is not present or has an invalid value, that piece is ignored by the pipeline.

Tokens info with a list of tokens and the corresponding list of token ids.

Used in: ComputeTokensResponse

repeated bytes tokens = 1
A list of tokens from the input.
repeated int64 token_ids = 2
A list of token ids from the input.
string role = 3
Optional. Optional fields for the role from the corresponding Content.

Tool details that the model may use to generate response. A `Tool` is a piece of code that enables the system to interact with external systems to perform an action, or set of actions, outside of knowledge and scope of the model. A Tool object should contain exactly one type of Tool (e.g FunctionDeclaration, Retrieval or GoogleSearchRetrieval).

Used in: CachedContent, CountTokensRequest, GenerateContentRequest

repeated FunctionDeclaration function_declarations = 1
Optional. Function tool type. One or more function declarations to be passed to the model along with the current user query. Model may decide to call a subset of these functions by populating [FunctionCall][google.cloud.aiplatform.v1.Part.function_call] in the response. User should provide a [FunctionResponse][google.cloud.aiplatform.v1.Part.function_response] for each function call in the next turn. Based on the function responses, Model will generate the final response back to the user. Maximum 128 function declarations can be provided.
optional Retrieval retrieval = 2
Optional. Retrieval tool type. System will always execute the provided retrieval tool(s) to get external knowledge to answer the prompt. Retrieval results are presented to the model for generation.
optional Tool.GoogleSearch google_search = 7
Optional. GoogleSearch tool type. Tool to support Google Search in Model. Powered by Google.
optional GoogleSearchRetrieval google_search_retrieval = 3
Optional. GoogleSearchRetrieval tool type. Specialized retrieval tool that is powered by Google search.
optional EnterpriseWebSearch enterprise_web_search = 6
Optional. Tool to support searching public web data, powered by Vertex AI Search and Sec4 compliance.
optional Tool.CodeExecution code_execution = 4
Optional. CodeExecution tool type. Enables the model to execute code as part of generation.

Tool that executes code generated by the model, and automatically returns the result to the model. See also [ExecutableCode]and [CodeExecutionResult] which are input and output to this tool.

Used in: Tool

(message has no fields)

GoogleSearch tool type. Tool to support Google Search in Model. Powered by Google.

Used in: Tool

(message has no fields)

Input for tool call valid metric.

optional ToolCallValidSpec metric_spec = 1
Required. Spec for tool call valid metric.
repeated ToolCallValidInstance instances = 2
Required. Repeated tool call valid instances.

Spec for tool call valid instance.

Used in: ToolCallValidInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Required. Ground truth used to compare against the prediction.

Tool call valid metric value for an instance.

Used in: ToolCallValidResults

optional float score = 1
Output only. Tool call valid score.

Results for tool call valid metric.

repeated ToolCallValidMetricValue tool_call_valid_metric_values = 1
Output only. Tool call valid metric values.

Spec for tool call valid metric.

Used in: ToolCallValidInput

(message has no fields)

Tool config. This config is shared for all tools provided in the request.

Used in: CachedContent, GenerateContentRequest

optional FunctionCallingConfig function_calling_config = 1
Optional. Function calling config.
optional RetrievalConfig retrieval_config = 2
Optional. Retrieval config.

Input for tool name match metric.

optional ToolNameMatchSpec metric_spec = 1
Required. Spec for tool name match metric.
repeated ToolNameMatchInstance instances = 2
Required. Repeated tool name match instances.

Spec for tool name match instance.

Used in: ToolNameMatchInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Required. Ground truth used to compare against the prediction.

Tool name match metric value for an instance.

Used in: ToolNameMatchResults

optional float score = 1
Output only. Tool name match score.

Results for tool name match metric.

repeated ToolNameMatchMetricValue tool_name_match_metric_values = 1
Output only. Tool name match metric values.

Spec for tool name match metric.

Used in: ToolNameMatchInput

(message has no fields)

Input for tool parameter key value match metric.

optional ToolParameterKVMatchSpec metric_spec = 1
Required. Spec for tool parameter key value match metric.
repeated ToolParameterKVMatchInstance instances = 2
Required. Repeated tool parameter key value match instances.

Spec for tool parameter key value match instance.

Used in: ToolParameterKVMatchInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Required. Ground truth used to compare against the prediction.

Tool parameter key value match metric value for an instance.

Used in: ToolParameterKVMatchResults

optional float score = 1
Output only. Tool parameter key value match score.

Results for tool parameter key value match metric.

repeated ToolParameterKVMatchMetricValue tool_parameter_kv_match_metric_values = 1
Output only. Tool parameter key value match metric values.

Spec for tool parameter key value match metric.

Used in: ToolParameterKVMatchInput

bool use_strict_string_match = 1
Optional. Whether to use STRICT string match on parameter values.

Input for tool parameter key match metric.

optional ToolParameterKeyMatchSpec metric_spec = 1
Required. Spec for tool parameter key match metric.
repeated ToolParameterKeyMatchInstance instances = 2
Required. Repeated tool parameter key match instances.

Spec for tool parameter key match instance.

Used in: ToolParameterKeyMatchInput

optional string prediction = 1
Required. Output of the evaluated model.
optional string reference = 2
Required. Ground truth used to compare against the prediction.

Tool parameter key match metric value for an instance.

Used in: ToolParameterKeyMatchResults

optional float score = 1
Output only. Tool parameter key match score.

Results for tool parameter key match metric.