openapi: 3.2.0
info:
  title: Sensei Machine Learning ML Services API
  description: "Sensei Machine Learning API provides a mechanism for data scientists to organize and manage ML services from algorithm onboarding through experimentation and to service deployment.\n\nRelated documentation:\n  * [Data Science Workspace documentation](http://www.adobe.com/go/data-science-overview-en)\n  \nVisualize API calls with Postman (a free, third-party software):\n  * [Sensei Machine Learning API Postman collection on GitHub](https://github.com/adobe/experience-platform-postman-samples/blob/master/apis/experience-platform/Sensei%20Machine%20Learning%20API.postman_collection.json)\n  * [Video guide for creating the Postman environment](https://video.tv.adobe.com/v/28832)\n  * [Steps for importing environments and collections in Postman](https://learning.getpostman.com/docs/postman/collection_runs/using_environments_in_collection_runs/)\n\n\n**API paths**: \n  * PLATFORM Gateway URL: https://<span>platform.adobe.io/ \n  * Base path for this API: /data/sensei\n  * Example of a complete path: https://<span>platform.adobe.io/data/sensei/engines\n\n**Required headers**:\n  * All calls require the headers `Authorization`, `x-gw-ims-org-id`, and `x-api-key`. For more information on how to obtain these values, see the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).\n  * All resources in Experience Platform are isolated to specific virtual sandboxes. All requests to Platform APIs require the header `x-sandbox-name` whose value is the all-lowercase name of the sandbox the operation will take place in (for example, \"prod\"). See the [sandboxes overview](https://adobe.com/go/sandbox-overview-en) for more information.\n  * All requests with a payload in the request body (such as POST, PUT, and PATCH calls) must include the header `Content-Type` with a value of `application/json`. \n  \n**API error handling**: \n  - Refer to the Experience Platform API troubleshooting guide for [FAQs](https://experienceleague.adobe.com/docs/experience-platform/landing/troubleshooting.html#faq), [API status codes](https://experienceleague.adobe.com/docs/experience-platform/landing/troubleshooting.html#api-status-codes), and [request header errors](https://experienceleague.adobe.com/docs/experience-platform/landing/troubleshooting.html#request-header-errors)."
  version: '1.0'
servers:
- url: //{environment}.adobe.io/data/sensei
  variables:
    environment:
      default: platform
      enum:
      - platform
      - platform-stage
tags:
- name: MLServices
  description: An MLService is a published trained model that provides your organization with the ability to access and reuse previously developed models. A key feature of MLServices is the ability to automate training and scoring on a scheduled basis.
paths:
  /mlServices:
    get:
      tags:
      - MLServices
      summary: List all filtered MLServices
      description: MLServices may be searched by name or tags.
      operationId: retrieveMLServices
      parameters:
      - name: Authorization
        in: header
        description: The access token which can be copied from your Experience Platform integration, prefixed with "Bearer " (for example, `Bearer {ACCESS_TOKEN}`). For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-api-key
        in: header
        description: The API key which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-gw-ims-org-id
        in: header
        description: The Organization ID which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-sandbox-name
        in: header
        description: The name of the sandbox in which the operation will take place. See the [sandboxes overview documentation](https://adobe.com/go/sandbox-overview-en) for more information.
        required: true
        schema:
          type: string
      - name: start
        in: query
        description: Start value for pagination. Indicates the starting index for the items to return.
        schema:
          type: string
      - name: limit
        in: query
        description: Limit value for pagination. Indicates the requested number of items to return. The service may return a different number of items than requested.
        schema:
          maximum: 100
          minimum: 0
          type: integer
          format: int32
          default: 25
      - name: orderby
        in: query
        description: Sort order for pagination. Indicates the fields and direction to use for sorting in priority order.
        style: form
        explode: false
        schema:
          type: array
          items:
            type: string
          default: '["-updated"]'
      - name: property
        in: query
        description: Property matching expression for filtering paginated results. Indicates the comparison expression that items must match in order to be returned.
        style: form
        explode: false
        schema:
          type: array
          items:
            type: string
      responses:
        '200':
          x-summary: Success
          description: List of MLServices
          content:
            application/vnd.adobe.platform.sensei+json;profile=mlServiceListing.v1.json:
              schema:
                $ref: '#/components/schemas/MLServiceListing'
        '400':
          x-summary: Bad request
          description: Bad Request. The request is malformed.
          content: {}
        '401':
          x-summary: Unauthorized
          description: Unauthorized. The request has not been applied because it lacks valid authentication credentials for the target resource.
          content: {}
        '403':
          x-summary: Forbidden
          description: Forbidden. That combination of arguments is not allowed.
          content: {}
        '406':
          x-summary: Not acceptable
          description: Not Acceptable. Unacceptable content type. The caller sent an <b>Accept</b> header for a content type which does not exist on the server
          content: {}
        '500':
          x-summary: Internal server error
          description: There was an internal server error. Try the request again in a few minutes, and contact Adobe Support if the problem persists.
          content: {}
        '502':
          x-summary: Bad gateway
          description: Bad Gateway. This is an indicator that the search back-end did not send a valid response.
          content: {}
        503:
          x-summary: Service unavailable
          description: Service Unavailable. This is an indicator of a potential server overload.
          content: {}
        504:
          x-summary: Gateway timeout
          description: Gateway Timeout. This is an indicator that the search back-end did not complete a response within an allowed time period.
          content: {}
    post:
      tags:
      - MLServices
      summary: Create an MLService
      description: Creates a new MLService of an MLInstance
      operationId: createMLService
      parameters:
      - name: Authorization
        in: header
        description: The access token which can be copied from your Experience Platform integration, prefixed with "Bearer " (for example, `Bearer {ACCESS_TOKEN}`). For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-api-key
        in: header
        description: The API key which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-gw-ims-org-id
        in: header
        description: The Organization ID which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-sandbox-name
        in: header
        description: The name of the sandbox in which the operation will take place. See the [sandboxes overview documentation](https://adobe.com/go/sandbox-overview-en) for more information.
        required: true
        schema:
          type: string
      requestBody:
        description: MLService object that needs to be added. Note that a valid `mlInstanceId` has to be listed in the object.
        content:
          application/vnd.adobe.platform.sensei+json;profile=mlService.v1.json:
            schema:
              $ref: '#/components/schemas/MLService'
        required: true
      responses:
        201:
          x-summary: Success
          description: Creation successful
          headers:
            Location:
              description: URL indicating the location of the created MLService resource
              schema:
                type: string
          content:
            application/vnd.adobe.platform.sensei+json;profile=mlService.v1.json:
              schema:
                $ref: '#/components/schemas/MLService'
        '400':
          x-summary: Bad request
          description: Bad Request. The request is malformed.
          content: {}
        '401':
          x-summary: Unauthorized
          description: Unauthorized. The request has not been applied because it lacks valid authentication credentials for the target resource.
          content: {}
        '403':
          x-summary: Forbidden
          description: Forbidden. That combination of arguments is not allowed.
          content: {}
        '406':
          x-summary: Not acceptable
          description: Not Acceptable. Unacceptable content type. The caller sent an <b>Accept</b> header for a content type which does not exist on the server
          content: {}
        415:
          x-summary: Unsupported media type
          description: Unsupported Media Type. The server refuses to accept the request because the payload format is in an unsupported format.
          content: {}
        422:
          x-summary: Unprocessable entity
          description: Unprocessable Entity. Client sent a correct request that contained invalid values for one or more parameters.
          content: {}
        '500':
          x-summary: Internal server error
          description: There was an internal server error. Try the request again in a few minutes, and contact Adobe Support if the problem persists.
          content: {}
        '502':
          x-summary: Bad gateway
          description: Bad Gateway. This is an indicator that the search back-end did not send a valid response.
          content: {}
        503:
          x-summary: Service unavailable
          description: Service Unavailable. This is an indicator of a potential server overload.
          content: {}
        504:
          x-summary: Gateway timeout
          description: Gateway Timeout. This is an indicator that the search back-end did not complete a response within an allowed time period.
          content: {}
      x-codegen-request-body-name: mlService
    delete:
      tags:
      - MLServices
      summary: Delete all MLServices
      description: Deletes all MLServices for provided mlInstanceId
      operationId: deleteMLServices
      parameters:
      - name: Authorization
        in: header
        description: The access token which can be copied from your Experience Platform integration, prefixed with "Bearer " (for example, `Bearer {ACCESS_TOKEN}`). For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-api-key
        in: header
        description: The API key which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-gw-ims-org-id
        in: header
        description: The Organization ID which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-sandbox-name
        in: header
        description: The name of the sandbox in which the operation will take place. See the [sandboxes overview documentation](https://adobe.com/go/sandbox-overview-en) for more information.
        required: true
        schema:
          type: string
      - name: mlInstanceId
        in: query
        description: The mlInstanceId for which MLServices should be deleted
        required: true
        schema:
          type: string
      responses:
        '200':
          x-summary: Success
          description: Successful operation
          content:
            application/vnd.adobe.platform.sensei+json;profile=api-response.v1.json:
              schema:
                $ref: '#/components/schemas/SenseiApiResponse'
        '400':
          x-summary: Bad request
          description: Bad Request. The request is malformed.
          content: {}
        '401':
          x-summary: Unauthorized
          description: Unauthorized. The request has not been applied because it lacks valid authentication credentials for the target resource.
          content: {}
        404:
          x-summary: Not found
          description: The resource the client is looking for isn't there.
          content: {}
        410:
          x-summary: Resource gone
          description: Resource gone. The requested resource has been permanently removed.
          content: {}
        '500':
          x-summary: Internal server error
          description: There was an internal server error. Try the request again in a few minutes, and contact Adobe Support if the problem persists.
          content: {}
        503:
          x-summary: Service unavailable
          description: Service Unavailable. This is an indicator of a potential server overload.
          content: {}
  /mlServices/{serviceId}:
    get:
      tags:
      - MLServices
      summary: Retrieve an MLService
      operationId: retrieveMLService
      parameters:
      - name: Authorization
        in: header
        description: The access token which can be copied from your Experience Platform integration, prefixed with "Bearer " (for example, `Bearer {ACCESS_TOKEN}`). For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-api-key
        in: header
        description: The API key which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-gw-ims-org-id
        in: header
        description: The Organization ID which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-sandbox-name
        in: header
        description: The name of the sandbox in which the operation will take place. See the [sandboxes overview documentation](https://adobe.com/go/sandbox-overview-en) for more information.
        required: true
        schema:
          type: string
      - name: serviceId
        in: path
        description: The identifier of the Sensei Service
        required: true
        schema:
          type: string
      responses:
        '200':
          x-summary: Success
          description: MLService object
          content:
            ? ''
            : schema:
                $ref: '#/components/schemas/MLService'
            application/vnd.adobe.platform.sensei+json;profile=mlService.v1.json:
              schema:
                $ref: '#/components/schemas/MLService'
        '400':
          x-summary: Bad request
          description: Bad Request. The request is malformed.
          content: {}
        '401':
          x-summary: Unauthorized
          description: Unauthorized. The request has not been applied because it lacks valid authentication credentials for the target resource.
          content: {}
        '403':
          x-summary: Forbidden
          description: Forbidden. That combination of arguments is not allowed.
          content: {}
        404:
          x-summary: Not found
          description: The resource the client is looking for isn't there.
          content: {}
        '406':
          x-summary: Not acceptable
          description: Not Acceptable. Unacceptable content type. The caller sent an <b>Accept</b> header for a content type which does not exist on the server
          content: {}
        '500':
          x-summary: Internal server error
          description: There was an internal server error. Try the request again in a few minutes, and contact Adobe Support if the problem persists.
          content: {}
        '502':
          x-summary: Bad gateway
          description: Bad Gateway. This is an indicator that the search back-end did not send a valid response.
          content: {}
        503:
          x-summary: Service unavailable
          description: Service Unavailable. This is an indicator of a potential server overload.
          content: {}
        504:
          x-summary: Gateway timeout
          description: Gateway Timeout. This is an indicator that the search back-end did not complete a response within an allowed time period.
          content: {}
    put:
      tags:
      - MLServices
      summary: Update an MLService
      operationId: updateMLService
      parameters:
      - name: Authorization
        in: header
        description: The access token which can be copied from your Experience Platform integration, prefixed with "Bearer " (for example, `Bearer {ACCESS_TOKEN}`). For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-api-key
        in: header
        description: The API key which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-gw-ims-org-id
        in: header
        description: The Organization ID which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-sandbox-name
        in: header
        description: The name of the sandbox in which the operation will take place. See the [sandboxes overview documentation](https://adobe.com/go/sandbox-overview-en) for more information.
        required: true
        schema:
          type: string
      - name: serviceId
        in: path
        description: The identifier of the Sensei Service
        required: true
        schema:
          type: string
      - name: If-Match
        in: header
        description: Optional header field with ETag value provided by client
        schema:
          type: string
      requestBody:
        description: MLService object to be updated
        content:
          application/vnd.adobe.platform.sensei+json;profile=mlService.v1.json:
            schema:
              $ref: '#/components/schemas/MLService'
        required: true
      responses:
        '200':
          x-summary: Success
          description: MLService object
          content:
            application/vnd.adobe.platform.sensei+json;profile=mlService.v1.json:
              schema:
                $ref: '#/components/schemas/MLService'
        '400':
          x-summary: Bad request
          description: Bad Request. The request is malformed.
          content: {}
        '401':
          x-summary: Unauthorized
          description: Unauthorized. The request has not been applied because it lacks valid authentication credentials for the target resource.
          content: {}
        '403':
          x-summary: Forbidden
          description: Forbidden. That combination of arguments is not allowed.
          content: {}
        404:
          x-summary: Not found
          description: The resource the client is looking for isn't there.
          content: {}
        '406':
          x-summary: Not acceptable
          description: Not Acceptable. Unacceptable content type. The caller sent an <b>Accept</b> header for a content type which does not exist on the server
          content: {}
        '500':
          x-summary: Internal server error
          description: There was an internal server error. Try the request again in a few minutes, and contact Adobe Support if the problem persists.
          content: {}
        '502':
          x-summary: Bad gateway
          description: Bad Gateway. This is an indicator that the search back-end did not send a valid response.
          content: {}
        503:
          x-summary: Service unavailable
          description: Service Unavailable. This is an indicator of a potential server overload.
          content: {}
        504:
          x-summary: Gateway timeout
          description: Gateway Timeout. This is an indicator that the search back-end did not complete a response within an allowed time period.
          content: {}
      x-codegen-request-body-name: mlService
    delete:
      tags:
      - MLServices
      summary: Delete an MLService
      operationId: deleteMLService
      parameters:
      - name: Authorization
        in: header
        description: The access token which can be copied from your Experience Platform integration, prefixed with "Bearer " (for example, `Bearer {ACCESS_TOKEN}`). For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-api-key
        in: header
        description: The API key which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-gw-ims-org-id
        in: header
        description: The Organization ID which can be copied from your Experience Platform integration. For more information on how to obtain this value, visit the documentation covering making API calls in the [authentication tutorial](http://www.adobe.com/go/platform-api-authentication-en).
        required: true
        schema:
          type: string
      - name: x-sandbox-name
        in: header
        description: The name of the sandbox in which the operation will take place. See the [sandboxes overview documentation](https://adobe.com/go/sandbox-overview-en) for more information.
        required: true
        schema:
          type: string
      - name: serviceId
        in: path
        description: The identifier of the Sensei Service
        required: true
        schema:
          type: string
      responses:
        '200':
          x-summary: Success
          description: MLService deleted successfully
          content:
            application/vnd.adobe.platform.sensei+json;profile=api-response.v1.json:
              schema:
                $ref: '#/components/schemas/SenseiApiResponse'
        '400':
          x-summary: Bad request
          description: Bad Request. The request is malformed.
          content: {}
        '401':
          x-summary: Unauthorized
          description: Unauthorized. The request has not been applied because it lacks valid authentication credentials for the target resource.
          content: {}
        '403':
          x-summary: Forbidden
          description: Forbidden. That combination of arguments is not allowed.
          content: {}
        404:
          x-summary: Not found
          description: The resource the client is looking for isn't there.
          content: {}
        '406':
          x-summary: Not acceptable
          description: Not Acceptable. Unacceptable content type. The caller sent an <b>Accept</b> header for a content type which does not exist on the server
          content: {}
        '500':
          x-summary: Internal server error
          description: There was an internal server error. Try the request again in a few minutes, and contact Adobe Support if the problem persists.
          content: {}
        '502':
          x-summary: Bad gateway
          description: Bad Gateway. This is an indicator that the search back-end did not send a valid response.
          content: {}
        503:
          x-summary: Service unavailable
          description: Service Unavailable. This is an indicator of a potential server overload.
          content: {}
        504:
          x-summary: Gateway timeout
          description: Gateway Timeout. This is an indicator that the search back-end did not complete a response within an allowed time period.
          content: {}
components:
  schemas:
    LinkObject:
      required:
      - href
      type: object
      properties:
        href:
          type: string
          description: Either an URI (RFC3986) or an URI Template (RFC6570) in which case the linkObject SHOULD have a 'templated' attribute whose value is true
        templated:
          type: boolean
          description: True if the link href is templated, false otherwise which is also the default
          default: false
        type:
          type: string
          description: A hint to indicate the media type expected when dereferencing the target resource
        method:
          type: string
          description: The HTTP method to be applied when using this link. This should be a valid HTTP method name such as GET, PUT, POST, PATCH, DELETE where GET implies support for HEAD as well
          default: GET
        title:
          type: string
          description: A label with a human-readable identifier (as defined by RFC5988)
        hrefLang:
          type: string
          description: The language of the target resource (as defined by RFC5988)
    MLService:
      allOf:
      - $ref: '#/components/schemas/MLServiceSummary'
      - type: object
        properties:
          tags:
            type: object
            additionalProperties:
              type: string
          _links:
            type: object
            properties:
              self:
                $ref: '#/components/schemas/LinkObject'
              getMLService:
                $ref: '#/components/schemas/LinkObject'
              deleteMLService:
                $ref: '#/components/schemas/LinkObject'
              getMLInstance:
                $ref: '#/components/schemas/LinkObject'
    MLServiceListingLinks:
      allOf:
      - $ref: '#/components/schemas/PagingNavigationAttributes'
      - type: object
        properties:
          self:
            $ref: '#/components/schemas/LinkObject'
          createMLInstance:
            $ref: '#/components/schemas/LinkObject'
    PagingNavigationAttributes:
      type: object
      properties:
        next:
          $ref: '#/components/schemas/LinkObject'
        page:
          $ref: '#/components/schemas/LinkObject'
    MLServiceListing:
      type: object
      properties:
        children:
          type: array
          items:
            $ref: '#/components/schemas/MLServiceSummary'
        _page:
          $ref: '#/components/schemas/PagingInformationAttributes'
        _links:
          $ref: '#/components/schemas/MLServiceListingLinks'
    SenseiApiResponse:
      type: object
      properties:
        type:
          type: string
          readOnly: true
          example: https://platform.adobe.io/data/foundation/ml/problems/illegal-argument
        title:
          type: string
          readOnly: true
          example: Illegal Argument
        status:
          type: integer
          format: int32
          readOnly: true
          example: 400
        detail:
          type: string
          readOnly: true
          example: Model specification name must not be empty
        additionalDetails:
          type: object
          additionalProperties:
            type: string
    MLServiceSummary:
      required:
      - mlInstanceId
      - name
      - scoringDataSetId
      - scoringSchedule
      - scoringTimeframe
      type: object
      properties:
        id:
          type: string
          readOnly: true
        name:
          type: string
        description:
          type: string
        mlInstanceId:
          type: string
        trainingExperimentId:
          type: string
        trainingExperimentRunId:
          type: string
        trainingDataSetId:
          type: string
        trainingTimeframe:
          type: integer
          format: int32
        scoringExperimentId:
          type: string
        scoringDataSetId:
          type: string
        scoringTimeframe:
          type: integer
          format: int32
        trainingSchedule:
          type: object
          properties:
            startTime:
              type: string
              format: datetime
            endTime:
              type: string
              format: datetime
            cron:
              type: string
        scoringSchedule:
          type: object
          properties:
            startTime:
              type: string
              format: datetime
            endTime:
              type: string
              format: datetime
            cron:
              type: string
        created:
          type: string
          format: date-time
          readOnly: true
        createdBy:
          type: object
          properties:
            displayName:
              type: string
            avatar:
              type: string
            userId:
              type: string
        updated:
          type: string
          format: date-time
          readOnly: true
        deleted:
          type: boolean
          default: false
        _links:
          type: object
          properties:
            self:
              $ref: '#/components/schemas/LinkObject'
    PagingInformationAttributes:
      type: object
      properties:
        orderby:
          type: string
          readOnly: true
        property:
          type: string
          readOnly: true
        start:
          type: string
          readOnly: true
        totalCount:
          type: integer
          format: int64
          readOnly: true
        count:
          type: integer
          format: int64
          readOnly: true
        next:
          type: string
          readOnly: true
x-original-swagger-version: '2.0'