kpt-grpc-demo/google/cloud/aiplatform/v1beta1/model_deployment_monitoring_job.proto

// Copyright 2022 Google LLC
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

syntax = "proto3";

package google.cloud.aiplatform.v1beta1;

import "google/api/field_behavior.proto";
import "google/api/resource.proto";
import "google/cloud/aiplatform/v1beta1/encryption_spec.proto";
import "google/cloud/aiplatform/v1beta1/feature_monitoring_stats.proto";
import "google/cloud/aiplatform/v1beta1/io.proto";
import "google/cloud/aiplatform/v1beta1/job_state.proto";
import "google/cloud/aiplatform/v1beta1/model_monitoring.proto";
import "google/protobuf/duration.proto";
import "google/protobuf/struct.proto";
import "google/protobuf/timestamp.proto";
import "google/rpc/status.proto";

option csharp_namespace = "Google.Cloud.AIPlatform.V1Beta1";
option go_package = "google.golang.org/genproto/googleapis/cloud/aiplatform/v1beta1;aiplatform";
option java_multiple_files = true;
option java_outer_classname = "ModelDeploymentMonitoringJobProto";
option java_package = "com.google.cloud.aiplatform.v1beta1";
option php_namespace = "Google\\Cloud\\AIPlatform\\V1beta1";
option ruby_package = "Google::Cloud::AIPlatform::V1beta1";

// The Model Monitoring Objective types.
enum ModelDeploymentMonitoringObjectiveType {
  // Default value, should not be set.
  MODEL_DEPLOYMENT_MONITORING_OBJECTIVE_TYPE_UNSPECIFIED = 0;

  // Raw feature values' stats to detect skew between Training-Prediction
  // datasets.
  RAW_FEATURE_SKEW = 1;

  // Raw feature values' stats to detect drift between Serving-Prediction
  // datasets.
  RAW_FEATURE_DRIFT = 2;

  // Feature attribution scores to detect skew between Training-Prediction
  // datasets.
  FEATURE_ATTRIBUTION_SKEW = 3;

  // Feature attribution scores to detect skew between Prediction datasets
  // collected within different time windows.
  FEATURE_ATTRIBUTION_DRIFT = 4;
}

// 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.
message ModelDeploymentMonitoringJob {
  option (google.api.resource) = {
    type: "aiplatform.googleapis.com/ModelDeploymentMonitoringJob"
    pattern: "projects/{project}/locations/{location}/modelDeploymentMonitoringJobs/{model_deployment_monitoring_job}"
  };

  // All metadata of most recent monitoring pipelines.
  message LatestMonitoringPipelineMetadata {
    // The time that most recent monitoring pipelines that is related to this
    // run.
    google.protobuf.Timestamp run_time = 1;

    // The status of the most recent monitoring pipeline.
    google.rpc.Status status = 2;
  }

  // The state to Specify the monitoring pipeline.
  enum MonitoringScheduleState {
    // Unspecified state.
    MONITORING_SCHEDULE_STATE_UNSPECIFIED = 0;

    // The pipeline is picked up and wait to run.
    PENDING = 1;

    // The pipeline is offline and will be scheduled for next run.
    OFFLINE = 2;

    // The pipeline is running.
    RUNNING = 3;
  }

  // Output only. Resource name of a ModelDeploymentMonitoringJob.
  string name = 1 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Required. The user-defined name of the ModelDeploymentMonitoringJob.
  // The name can be up to 128 characters long and can be consist of any UTF-8
  // characters.
  // Display name of a ModelDeploymentMonitoringJob.
  string display_name = 2 [(google.api.field_behavior) = REQUIRED];

  // Required. Endpoint resource name.
  // Format: `projects/{project}/locations/{location}/endpoints/{endpoint}`
  string endpoint = 3 [
    (google.api.field_behavior) = REQUIRED,
    (google.api.resource_reference) = {
      type: "aiplatform.googleapis.com/Endpoint"
    }
  ];

  // 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'.
  JobState state = 4 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Output only. Schedule state when the monitoring job is in Running state.
  MonitoringScheduleState schedule_state = 5 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Output only. Latest triggered monitoring pipeline metadata.
  LatestMonitoringPipelineMetadata latest_monitoring_pipeline_metadata = 25 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Required. The config for monitoring objectives. This is a per DeployedModel config.
  // Each DeployedModel needs to be configured separately.
  repeated ModelDeploymentMonitoringObjectiveConfig model_deployment_monitoring_objective_configs = 6 [(google.api.field_behavior) = REQUIRED];

  // Required. Schedule config for running the monitoring job.
  ModelDeploymentMonitoringScheduleConfig model_deployment_monitoring_schedule_config = 7 [(google.api.field_behavior) = REQUIRED];

  // Required. Sample Strategy for logging.
  SamplingStrategy logging_sampling_strategy = 8 [(google.api.field_behavior) = REQUIRED];

  // Alert config for model monitoring.
  ModelMonitoringAlertConfig model_monitoring_alert_config = 15;

  // 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.
  string predict_instance_schema_uri = 9;

  // Sample Predict instance, same format as [PredictRequest.instances][google.cloud.aiplatform.v1beta1.PredictRequest.instances],
  // this can be set as a replacement of
  // [ModelDeploymentMonitoringJob.predict_instance_schema_uri][google.cloud.aiplatform.v1beta1.ModelDeploymentMonitoringJob.predict_instance_schema_uri]. If not set,
  // we will generate predict schema from collected predict requests.
  google.protobuf.Value sample_predict_instance = 19;

  // 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.v1beta1.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.
  string analysis_instance_schema_uri = 16;

  // 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
  repeated ModelDeploymentMonitoringBigQueryTable bigquery_tables = 10 [(google.api.field_behavior) = OUTPUT_ONLY];

  // 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.
  google.protobuf.Duration log_ttl = 17;

  // 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.
  map<string, string> labels = 11;

  // Output only. Timestamp when this ModelDeploymentMonitoringJob was created.
  google.protobuf.Timestamp create_time = 12 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Output only. Timestamp when this ModelDeploymentMonitoringJob was updated most recently.
  google.protobuf.Timestamp update_time = 13 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Output only. Timestamp when this monitoring pipeline will be scheduled to run for the
  // next round.
  google.protobuf.Timestamp next_schedule_time = 14 [(google.api.field_behavior) = OUTPUT_ONLY];

  // Stats anomalies base folder path.
  GcsDestination stats_anomalies_base_directory = 20;

  // 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.
  EncryptionSpec encryption_spec = 21;

  // 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).
  bool enable_monitoring_pipeline_logs = 22;

  // Output only. Only populated when the job's state is `JOB_STATE_FAILED` or
  // `JOB_STATE_CANCELLED`.
  google.rpc.Status error = 23 [(google.api.field_behavior) = OUTPUT_ONLY];
}

// ModelDeploymentMonitoringBigQueryTable specifies the BigQuery table name
// as well as some information of the logs stored in this table.
message ModelDeploymentMonitoringBigQueryTable {
  // Indicates where does the log come from.
  enum LogSource {
    // Unspecified source.
    LOG_SOURCE_UNSPECIFIED = 0;

    // Logs coming from Training dataset.
    TRAINING = 1;

    // Logs coming from Serving traffic.
    SERVING = 2;
  }

  // Indicates what type of traffic does the log belong to.
  enum LogType {
    // Unspecified type.
    LOG_TYPE_UNSPECIFIED = 0;

    // Predict logs.
    PREDICT = 1;

    // Explain logs.
    EXPLAIN = 2;
  }

  // The source of log.
  LogSource log_source = 1;

  // The type of log.
  LogType log_type = 2;

  // 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 bigquery_table_path = 3;
}

// ModelDeploymentMonitoringObjectiveConfig contains the pair of
// deployed_model_id to ModelMonitoringObjectiveConfig.
message ModelDeploymentMonitoringObjectiveConfig {
  // The DeployedModel ID of the objective config.
  string deployed_model_id = 1;

  // The objective config of for the modelmonitoring job of this deployed model.
  ModelMonitoringObjectiveConfig objective_config = 2;
}

// The config for scheduling monitoring job.
message ModelDeploymentMonitoringScheduleConfig {
  // 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.
  google.protobuf.Duration monitor_interval = 1 [(google.api.field_behavior) = REQUIRED];

  // 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.v1beta1.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.
  google.protobuf.Duration monitor_window = 2;
}

// Statistics and anomalies generated by Model Monitoring.
message ModelMonitoringStatsAnomalies {
  // Historical Stats (and Anomalies) for a specific Feature.
  message FeatureHistoricStatsAnomalies {
    // Display Name of the Feature.
    string feature_display_name = 1;

    // Threshold for anomaly detection.
    ThresholdConfig threshold = 3;

    // Stats calculated for the Training Dataset.
    FeatureStatsAnomaly training_stats = 4;

    // A list of historical stats generated by different time window's
    // Prediction Dataset.
    repeated FeatureStatsAnomaly prediction_stats = 5;
  }

  // Model Monitoring Objective those stats and anomalies belonging to.
  ModelDeploymentMonitoringObjectiveType objective = 1;

  // Deployed Model ID.
  string deployed_model_id = 2;

  // Number of anomalies within all stats.
  int32 anomaly_count = 3;

  // A list of historical Stats and Anomalies generated for all Features.
  repeated FeatureHistoricStatsAnomalies feature_stats = 4;
}