openapi: 3.0.0 info: description: 'API to interact and manage the lifecycle of your machine learning models deployed through Seldon Deploy.' title: Seldon Deploy AlertingService MetricsServer API version: v1alpha1 servers: - url: http://X.X.X.X/seldon-deploy/api/v1alpha1 - url: https://X.X.X.X/seldon-deploy/api/v1alpha1 security: - OAuth2: - '[]' tags: - description: Metrics Server APIs for Seldon Deploy name: MetricsServer paths: /namespaces/{namespace}/seldondeployments/{name}/monitor/metrics-server: get: description: Read the specified Seldon Deployment Metrics Server operationId: ListMetricsServerSeldonDeployment parameters: - description: Name identifies a resource in: path name: name required: true schema: type: string - description: Namespace provides a logical grouping of resources in: path name: namespace required: true schema: type: string responses: '200': $ref: '#/components/responses/DetectorListResponse' '500': $ref: '#/components/responses/ErrorDefault' tags: - MetricsServer post: description: Create the specified Seldon Deployment Metrics Server operationId: CreateMetricsServerSeldonDeployment parameters: - description: Name identifies a resource in: path name: name required: true schema: type: string - description: Namespace provides a logical grouping of resources in: path name: namespace required: true schema: type: string requestBody: $ref: '#/components/requestBodies/DetectorConfigData' responses: '200': $ref: '#/components/responses/DetectorResponse' '400': $ref: '#/components/responses/ErrorDefault' '500': $ref: '#/components/responses/ErrorDefault' tags: - MetricsServer /namespaces/{namespace}/seldondeployments/{name}/monitor/metrics-server/{detectorName}: delete: description: Read the specified Seldon Deployment Metrics Server operationId: DeleteMetricsServerSeldonDeployment parameters: - description: Name identifies a resource in: path name: name required: true schema: type: string - description: Namespace provides a logical grouping of resources in: path name: namespace required: true schema: type: string - description: Detector Name in: path name: detectorName required: true schema: type: string responses: '200': $ref: '#/components/responses/MessageResponse' '500': $ref: '#/components/responses/ErrorDefault' tags: - MetricsServer get: description: Read the specified Seldon Deployment Metrics Server operationId: ReadMetricsServerSeldonDeployment parameters: - description: Name identifies a resource in: path name: name required: true schema: type: string - description: Namespace provides a logical grouping of resources in: path name: namespace required: true schema: type: string - description: Detector Name in: path name: detectorName required: true schema: type: string responses: '200': $ref: '#/components/responses/DetectorResponse' '500': $ref: '#/components/responses/ErrorDefault' tags: - MetricsServer components: schemas: DetectorData: properties: config: $ref: '#/components/schemas/DetectorConfiguration' deployment_kind: type: string deployment_name: type: string detector_type: type: string name: type: string namespace: type: string status: $ref: '#/components/schemas/DetectorStatus' type: object BasicDetectorConfiguration: properties: drift_batch_size: type: string drift_type_feature: type: boolean env_secret_ref: type: string project: type: string storage_uri: type: string type: object Error: description: Error Error properties: code: description: Code format: int64 type: integer message: description: Message type: string requestId: description: RequestId type: string type: object DetectorDeploymentConfiguration: properties: event_source: type: string event_type: type: string http_port: type: string memory_requirement: type: string model_name: type: string prom_scraping: type: boolean protocol: type: string description: 'For model inference, Seldon recommends using the industry-standard Open Inference Protocol (OIP) as the preferred protocol over others.
The corresponding protocol OIP value for Seldon Deployment detectors is ''kfserving.http''. For Seldon ML Pipelines, this is the only supported protocol.
For more information, please refer to the Seldon documentation: https://docs.seldon.ai/seldon-core-2/apis/inference/v2' reply_url: type: string user_permission: format: int64 type: integer type: object DetectorConfiguration: properties: basic: $ref: '#/components/schemas/BasicDetectorConfiguration' deployment: $ref: '#/components/schemas/DetectorDeploymentConfiguration' type: object DetectorStatus: type: string DetectorConfigData: properties: config: $ref: '#/components/schemas/DetectorConfiguration' name: type: string type: object Message: description: Message Message properties: code: description: Code format: int64 type: integer message: description: Message type: string type: object responses: DetectorListResponse: description: Detector configuration data list content: application/json: schema: items: $ref: '#/components/schemas/DetectorData' type: array DetectorResponse: description: Detector configuration data content: application/json: schema: $ref: '#/components/schemas/DetectorData' MessageResponse: description: GenericMessage Message content: application/json: schema: $ref: '#/components/schemas/Message' ErrorDefault: description: ErrorDefault Error content: application/json: schema: $ref: '#/components/schemas/Error' requestBodies: DetectorConfigData: content: application/json: schema: $ref: '#/components/schemas/DetectorConfigData' description: Deployment Detector Data required: true securitySchemes: OAuth2: type: oauth2 flows: password: tokenUrl: https://Y.Y.Y.Y scopes: email: '' groups: '' openid: '' profile: ''