swagger: '2.0'
info:
title: Microsoft Azure Computer Vision Client
description: >-
The Computer Vision API provides state-of-the-art algorithms to process
images and return information. For example, it can be used to extract text
using Read OCR, caption an image using descriptive natural language, detect
objects, people, and more.
version: 2023-04-01-preview
paths:
/imageanalysis:segment:
post:
tags:
- Operations
summary: >-
Microsoft Azure Analyze The Input Image The Request Either Contains An Image Stream With Any Content Type [ Image * , Application Octet Stream ], Or A Json Payload Which Includes A Url Property To Be Used To Retrieve The Image Stream An Image Stream Of Content Type Image Png Is Returned, Where The Pixel Values Depend On The Analysis Mode The Returned Image Has The Same Dimensions As The Input Image For Modes: Foregroundmatting The Returned Image Has The Same Aspect Ratio And Same Dimensions As The Input Image Up To A Limit Of 16 Megapixels For Modes: Backgroundremoval
operationId: microsoftAzureImageanalysisSegment
consumes:
- application/json
produces:
- image/png
- application/json
parameters:
- in: query
name: mode
description: The analysis mode requested.
type: string
enum:
- backgroundRemoval
- foregroundMatting
x-ms-enum:
name: SegmentationMode
modelAsString: true
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: >-
A JSON document with a URL pointing to the image that is to be
analyzed.
required: true
schema:
$ref: '#/definitions/ImageUrl'
responses:
'200':
description: Success
schema:
type: file
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Segment_BackgroundRemoval_FromImageUrl:
$ref: ./examples/Segment_BackgroundRemoval_FromImageUrl.json
Segment_ForegroundMatting_FromImageUrl:
$ref: ./examples/Segment_ForegroundMatting_FromImageUrl.json
description: Needs a more full description created.
/retrieval:vectorizeText:
post:
tags:
- ImageRetrieval
summary: 'Microsoft Azure Return Vector From A Text'
operationId: microsoftAzureImageretrievalVectorizetext
consumes:
- application/json
produces:
- application/json
parameters:
- in: query
name: model-version
description: Model version.
type: string
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: Request of VectorizeText.
schema:
$ref: '#/definitions/VectorizeTextRequestApiModel'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/SingleVectorResultApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ImageRetrieval_VectorizeText:
$ref: ./examples/ImageRetrieval_VectorizeText.json
description: Needs a more full description created.
/retrieval:vectorizeImage:
post:
tags:
- ImageRetrieval
summary: 'Microsoft Azure Return Vector From An Image'
operationId: microsoftAzureImageretrievalVectorizeimage
consumes:
- application/json
produces:
- application/json
parameters:
- in: query
name: model-version
description: Model version.
type: string
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: >-
A JSON document with a URL pointing to the image that is to be
analyzed.
required: true
schema:
$ref: '#/definitions/ImageUrl'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/SingleVectorResultApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ImageRetrieval_VectorizeImage:
$ref: ./examples/ImageRetrieval_VectorizeImage.json
description: Needs a more full description created.
/imagecomposition:stitch:
post:
tags:
- Operations
summary: 'Microsoft Azure Run The Image Stitching Operation Against A Sequence Of Images'
operationId: microsoftAzureImagecompositionStitch
consumes:
- application/json-patch+json
produces:
- image/jpeg
- application/json
parameters:
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: Input images urls that pass into image stitching operation.
required: true
schema:
$ref: '#/definitions/ImageStitchingRequestApiModel'
responses:
'200':
description: Success
schema:
type: file
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ImageComposition_Stitch:
$ref: ./examples/ImageComposition_Stitch.json
description: Needs a more full description created.
/imagecomposition:rectify:
post:
tags:
- Operations
summary: >-
Microsoft Azure Run The Image Rectification Operation Against An Image With 4 Control Points Provided In The Parameter
operationId: microsoftAzureImagecompositionRectify
consumes:
- application/json-patch+json
produces:
- image/jpeg
- application/json
parameters:
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: >-
Input image url and control points that are passed into
rectification operation.
required: true
schema:
$ref: '#/definitions/ImageRectificationRequestApiModel'
responses:
'200':
description: Success
schema:
type: file
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ImageComposition_Rectify:
$ref: ./examples/ImageComposition_Rectify.json
description: Needs a more full description created.
/planogramcompliance:match:
post:
tags:
- Operations
summary: >-
Microsoft Azure Run The Planogram Matching Operation Against A Planogram And A Product Understanding Result
operationId: microsoftAzurePlanogramcomplianceMatch
consumes:
- application/json-patch+json
produces:
- application/json
parameters:
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: Input to pass into the planogram matching operation.
schema:
$ref: '#/definitions/PlanogramMatchingRequestApiModel'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/PlanogramMatchingResultApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
PlanogramCompliance_Match:
$ref: ./examples/PlanogramCompliance_Match.json
description: Needs a more full description created.
/productrecognition/{modelName}/runs/{runName}:
put:
tags:
- Operations
summary: 'Microsoft Azure Run The Product Recognition Against A Model With An Image'
operationId: microsoftAzureProductrecognitionCreate
produces:
- application/json
parameters:
- in: path
name: modelName
description: The name of the model to run product recognition with.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: path
name: runName
description: The name of the product recognition run.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
responses:
'201':
description: Created
schema:
$ref: '#/definitions/ProductRecognitionApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ProductRecognition_Create:
$ref: ./examples/ProductRecognition_Create.json
description: Needs a more full description created.
get:
tags:
- Operations
summary: 'Microsoft Azure Get Information About A Specific Product Recognition Run'
operationId: microsoftAzureProductrecognitionGet
produces:
- application/json
parameters:
- in: path
name: modelName
description: The name of the model the product recognition run belongs to.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: path
name: runName
description: The name of the product recognition run.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/ProductRecognitionApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ProductRecognition_Get:
$ref: ./examples/ProductRecognition_Get.json
description: Needs a more full description created.
delete:
tags:
- Operations
summary: "Microsoft Azure Delete A Product Recognition Run A Product Recognition Run Can Be Deleted If It Is In The Succeeded Or Failed States \r\n \r\nstatus Codes Returned:\r\n 204: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 404: A Product Recognition Run With The Specified Name Was Not Found"
operationId: microsoftAzureProductrecognitionDelete
produces:
- application/json
parameters:
- in: path
name: modelName
description: The name of the model to delete the product recognition run for.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: path
name: runName
description: The name of the product recognition run.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
responses:
'204':
description: No Content
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ProductRecognition_Delete:
$ref: ./examples/ProductRecognition_Delete.json
description: Needs a more full description created.
/productrecognition/{modelName}/runs:
get:
tags:
- Operations
summary: 'Microsoft Azure List All Product Recognition Run Of A Model'
operationId: microsoftAzureProductrecognitionList
produces:
- application/json
parameters:
- in: path
name: modelName
description: The name of the model the product recognition runs belongs to.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: query
name: skip
description: Number of product recognition runs to be skipped.
type: integer
format: int32
default: 0
maximum: 2147483647
minimum: 0
- in: query
name: top
description: >-
Number of product recognition runs to be returned after skipping.
The maximum allowed value is 30.
type: integer
format: int32
default: 10
maximum: 30
minimum: 1
- $ref: '#/parameters/ApiVersion'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/ProductRecognitionApiModelCollectionApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-pageable:
nextLinkName: nextLink
x-ms-examples:
ProductRecognition_List:
$ref: ./examples/ProductRecognition_List.json
description: Needs a more full description created.
/imageanalysis:analyze:
post:
tags:
- Operations
summary: >-
Microsoft Azure Analyze The Input Image The Request Either Contains Image Stream With Any Content Type [ Image * , Application Octet Stream ], Or A Json Payload Which Includes An Url Property To Be Used To Retrieve The Image Stream
operationId: microsoftAzureImageanalysisAnalyze
consumes:
- application/json
produces:
- application/json
parameters:
- in: query
name: features
description: >-
The visual features requested: tags, objects, caption,
denseCaptions, read, smartCrops, people. This parameter needs to be
specified if the parameter "model-name" is not specified.
type: array
items:
enum:
- tags
- caption
- denseCaptions
- objects
- read
- smartCrops
- people
type: string
x-ms-enum:
name: VisualFeature
modelAsString: true
collectionFormat: csv
- in: query
name: model-name
description: >-
The name of the custom trained model. This parameter needs to be
specified if the parameter "features" is not specified.
type: string
- in: query
name: language
description: >-
The desired language for output generation. If this parameter is not
specified, the default value is "en". See
https://aka.ms/cv-languages for a list of supported languages.
type: string
default: en
- in: query
name: smartcrops-aspect-ratios
description: >-
A list of aspect ratios to use for smartCrops feature. Aspect ratios
are calculated by dividing the target crop width by the height.
Supported values are between 0.75 and 1.8 (inclusive). Multiple
values should be comma-separated. If this parameter is not
specified, the service will return one crop suggestion with an
aspect ratio it sees fit between 0.5 and 2.0 (inclusive).
type: string
- in: query
name: gender-neutral-caption
description: >-
Boolean flag for enabling gender-neutral captioning for caption and
denseCaptions features. If this parameter is not specified, the
default value is "false".
type: boolean
default: false
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: >-
A JSON document with a URL pointing to the image that is to be
analyzed.
required: true
schema:
$ref: '#/definitions/ImageUrl'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/ImageAnalysisResult'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
AnalyzeImage_CustomModel:
$ref: ./examples/AnalyzeImage_CustomModel.json
description: Needs a more full description created.
/datasets/{name}:
put:
tags:
- Datasets
summary: "Microsoft Azure Register A New Dataset \r\n \r\nstatus Codes Returned:\r\n 201: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 409: A Dataset With The Specified Name Already Exists"
operationId: microsoftAzureDatasetsRegister
consumes:
- application/json-patch+json
produces:
- application/json
parameters:
- in: path
name: name
description: >-
A name that can be used to uniquely identify the dataset after it
has been registered.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: >-
Properties for the dataset, such as the list of URIs to the
annotation files.
required: true
schema:
$ref: '#/definitions/Dataset'
responses:
'201':
description: Created
schema:
$ref: '#/definitions/Dataset'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
RegisterDataset_ImageClassification:
$ref: ./examples/RegisterDataset_ImageClassification.json
description: Needs a more full description created.
get:
tags:
- Datasets
summary: "Microsoft Azure Get Information About A Specific Dataset \r\n \r\nstatus Codes Returned:\r\n 200: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 404: A Dataset With The Specified Name Was Not Found"
operationId: microsoftAzureDatasetsGet
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the dataset to get.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/Dataset'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Datasets_Get:
$ref: ./examples/Datasets_Get.json
description: Needs a more full description created.
patch:
tags:
- Datasets
summary: "Microsoft Azure Update The Properties Of An Existing Dataset \r\n \r\nstatus Codes Returned:\r\n 200: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 404: A Dataset With The Specified Name Was Not Found \r\n 412: An If Match Header Was Provided, But The Given Etag Did Not Match The Current Etag Value"
operationId: microsoftAzureDatasetsUpdate
consumes:
- application/json-patch+json
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the dataset to update.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: header
name: If-Match
description: "Optional ETag for the dataset to update.
\r\nIf an ETag is provided, then the dataset will be updated only if its current ETag value matches the given ETag.\r\nIf the ETag values don't match, then the update operation will fail with status code 412 (Precondition Failed).\r\nThis indicates that the dataset has been updated since the last time information for the dataset was obtained.
\r\nIf an ETag is not provided or its value is '*', then the dataset will always be updated regardless of the current ETag value."
type: string
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: Properties to update on the existing dataset.
required: true
schema:
$ref: '#/definitions/Dataset'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/Dataset'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Datasets_Update:
$ref: ./examples/Datasets_Update.json
description: Needs a more full description created.
delete:
tags:
- Datasets
summary: "Microsoft Azure Unregister A Dataset \r\n \r\nstatus Codes Returned:\r\n 204: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 404: A Dataset With The Specified Name Was Not Found \r\n 412: An If Match Header Was Provided, But The Given Etag Did Not Match The Current Etag Value"
operationId: microsoftAzureDatasetsUnregister
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the dataset to unregister.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: header
name: If-Match
description: "Optional ETag for the dataset to unregister.
\r\nIf an ETag is provided, then the dataset will be unregistered only if its current ETag value matches the given ETag.\r\nIf the ETag values don't match, then the unregister operation will fail with status code 412 (Precondition Failed).\r\nThis indicates that the dataset has been updated since the last time information for the dataset was obtained.
\r\nIf an ETag is not provided or its value is '*', then the dataset will always be unregistered regardless of the current ETag value."
type: string
- $ref: '#/parameters/ApiVersion'
responses:
'204':
description: No Content
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Datasets_Unregister:
$ref: ./examples/Datasets_Unregister.json
description: Needs a more full description created.
/datasets:
get:
tags:
- Datasets
summary: "Microsoft Azure Get A List Of Datasets That Have Been Registered \r\n \r\nstatus Codes Returned:\r\n 200: Operation Completed Successfully \r\n 400: The Request Was Malformed"
operationId: microsoftAzureDatasetsList
produces:
- application/json
parameters:
- in: query
name: skip
description: Number of datasets to be skipped.
type: integer
format: int32
default: 0
maximum: 2147483647
minimum: 0
- in: query
name: top
description: >-
Number of datasets to be returned after skipping. The maximum
allowed value is 30.
type: integer
format: int32
default: 10
maximum: 30
minimum: 1
- $ref: '#/parameters/ApiVersion'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/DatasetApiModelCollectionApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Datasets_List:
$ref: ./examples/Datasets_List.json
x-ms-pageable:
nextLinkName: nextLink
description: Needs a more full description created.
/models/{name}:
put:
tags:
- Models
summary: "Microsoft Azure Start Training A Custom Model \r\n \r\nstatus Codes Returned:\r\n 201: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 409: A Model With The Specified Name Already Exists"
operationId: microsoftAzureModelsCreate
consumes:
- application/json-patch+json
produces:
- application/json
parameters:
- in: path
name: name
description: >-
A name that can be used to uniquely identify the model after it has
been created.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: >-
Properties for the model, such as the name of the dataset to use to
train the model.
required: true
schema:
$ref: '#/definitions/Model'
responses:
'201':
description: Created
schema:
$ref: '#/definitions/Model'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Models_Create:
$ref: ./examples/Models_Create.json
description: Needs a more full description created.
get:
tags:
- Models
summary: "Microsoft Azure Get Information About A Specific Model \r\n \r\nstatus Codes Returned:\r\n 200: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 404: A Model With The Specified Name Was Not Found"
operationId: microsoftAzureModelsGet
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the model to get.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/Model'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Models_Get:
$ref: ./examples/Models_Get.json
description: Needs a more full description created.
delete:
tags:
- Models
summary: "Microsoft Azure Delete A Custom Model A Model Can Be Deleted If It Is In One Of The Succeeded , Failed , Or Canceled States \r\nif A Model Is In The Notstarted Or Training State, Cancel Training And Wait For Cancellation To Finish Before Deleting The Model \r\n \r\nstatus Codes Returned:\r\n 204: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 404: A Model With The Specified Name Was Not Found"
operationId: microsoftAzureModelsDelete
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the model to delete.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
responses:
'204':
description: No Content
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Models_Delete:
$ref: ./examples/Models_Delete.json
description: Needs a more full description created.
/models:
get:
tags:
- Models
summary: "Microsoft Azure Get A List Of The Available Models \r\n \r\nstatus Codes Returned:\r\n 200: Operation Completed Successfully \r\n 400: The Request Was Malformed"
operationId: microsoftAzureModelsList
produces:
- application/json
parameters:
- in: query
name: skip
description: Number of models to be skipped.
type: integer
format: int32
default: 0
maximum: 2147483647
minimum: 0
- in: query
name: top
description: >-
Number of models to be returned after skipping. The maximum allowed
value is 30.
type: integer
format: int32
default: 10
maximum: 30
minimum: 1
- $ref: '#/parameters/ApiVersion'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/ModelApiModelCollectionApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Models_List:
$ref: ./examples/Models_List.json
x-ms-pageable:
nextLinkName: nextLink
description: Needs a more full description created.
/models/{name}:cancel:
post:
tags:
- Models
summary: "Microsoft Azure Cancel Model Training \r\n \r\nstatus Codes Returned:\r\n 202: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 404: A Model With The Specified Name Was Not Found"
operationId: microsoftAzureModelsCanceltraining
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the model to cancel training.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
responses:
'202':
description: Accepted
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Models_CancelTraining:
$ref: ./examples/Models_CancelTraining.json
description: Needs a more full description created.
/models/{name}/evaluations/{evaluationName}:
put:
tags:
- Models
summary: "Microsoft Azure Evaluate An Existing Model \r\n \r\nstatus Codes Returned:\r\n 201: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 409: An Evaluation With The Specified Name Already Exists"
operationId: microsoftAzureModelevaluationsCreate
consumes:
- application/json-patch+json
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the model to evaluate.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: path
name: evaluationName
description: >-
A name that can be used to uniquely identify the evaluation after it
has been created.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: >-
Properties for the evaluation, such as the name of the dataset to
use to test the model.
required: true
schema:
$ref: '#/definitions/ModelEvaluation'
responses:
'201':
description: Created
schema:
$ref: '#/definitions/ModelEvaluation'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ModelEvaluations_Create:
$ref: ./examples/ModelEvaluations_Create.json
description: Needs a more full description created.
get:
tags:
- Models
summary: "Microsoft Azure Get Information About A Specific Model Evaluation \r\n \r\nstatus Codes Returned:\r\n 200: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 404: A Model Evaluation With The Specified Name Was Not Found"
operationId: microsoftAzureModelevaluationsGet
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the model to get the evaluation for.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: path
name: evaluationName
description: The name of the model evaluation to get.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/ModelEvaluation'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ModelEvaluations_Get:
$ref: ./examples/ModelEvaluations_Get.json
description: Needs a more full description created.
delete:
tags:
- Models
summary: "Microsoft Azure Delete A Model Evaluation A Model Evaluation Can Be Deleted If It Is In The Succeeded Or Failed States \r\n \r\nstatus Codes Returned:\r\n 204: Operation Completed Successfully \r\n 400: The Request Was Malformed \r\n 404: A Model Evaluation With The Specified Name Was Not Found"
operationId: microsoftAzureModelevaluationsDelete
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the model to delete the evaluation for.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: path
name: evaluationName
description: The name of the model evaluation to delete.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- $ref: '#/parameters/ApiVersion'
responses:
'204':
description: No Content
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ModelEvaluations_Delete:
$ref: ./examples/ModelEvaluations_Delete.json
description: Needs a more full description created.
/models/{name}/evaluations:
get:
tags:
- Models
summary: "Microsoft Azure Get A List Of The Available Evaluations For A Model \r\n \r\nstatus Codes Returned:\r\n 200: Operation Completed Successfully \r\n 400: The Request Was Malformed"
operationId: microsoftAzureModelevaluationsList
produces:
- application/json
parameters:
- in: path
name: name
description: The name of the model to get evaluations for.
required: true
type: string
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
- in: query
name: skip
description: Number of evaluations to be skipped.
type: integer
format: int32
default: 0
maximum: 2147483647
minimum: 0
- in: query
name: top
description: >-
Number of evaluations to be returned after skipping. The maximum
allowed value is 30.
type: integer
format: int32
default: 10
maximum: 30
minimum: 1
- $ref: '#/parameters/ApiVersion'
responses:
'200':
description: Success
schema:
$ref: '#/definitions/ModelEvaluationApiModelCollectionApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ModelEvaluations_List:
$ref: ./examples/ModelEvaluations_List.json
x-ms-pageable:
nextLinkName: nextLink
description: Needs a more full description created.
definitions:
AdultMatch:
description: An object describing adult content match.
required:
- confidence
- isMatch
type: object
properties:
isMatch:
description: A value indicating if the image is matched adult content.
type: boolean
confidence:
format: double
description: A value indicating the confidence level of matched adult content.
maximum: 1
minimum: 0
type: number
AdultResult:
description: >-
An object describing whether the image contains adult-oriented content
and/or is racy.
required:
- adult
- gore
- racy
type: object
properties:
adult:
$ref: '#/definitions/AdultMatch'
racy:
$ref: '#/definitions/AdultMatch'
gore:
$ref: '#/definitions/AdultMatch'
BoundingBox:
description: A bounding box for an area inside an image.
required:
- h
- w
- x
- 'y'
type: object
properties:
x:
format: int32
description: Left-coordinate of the top left point of the area, in pixels.
minimum: 0
type: integer
'y':
format: int32
description: Top-coordinate of the top left point of the area, in pixels.
minimum: 0
type: integer
w:
format: int32
description: Width measured from the top-left point of the area, in pixels.
minimum: 1
type: integer
h:
format: int32
description: Height measured from the top-left point of the area, in pixels.
minimum: 1
type: integer
CaptionResult:
description: A brief description of what the image depicts.
required:
- confidence
- text
type: object
properties:
text:
description: The text of the caption.
minLength: 1
type: string
confidence:
format: double
description: The level of confidence the service has in the caption.
maximum: 1
minimum: 0
type: number
CropRegion:
description: >-
A region identified for smart cropping. There will be one region returned
for each requested aspect ratio.
required:
- aspectRatio
- boundingBox
type: object
properties:
aspectRatio:
format: double
description: The aspect ratio of the crop region.
type: number
boundingBox:
$ref: '#/definitions/BoundingBox'
DenseCaption:
description: A brief description of what the image depicts.
required:
- confidence
- text
type: object
properties:
text:
description: The text of the caption.
minLength: 1
type: string
confidence:
format: double
description: The level of confidence the service has in the caption.
maximum: 1
minimum: 0
type: number
boundingBox:
$ref: '#/definitions/BoundingBox'
DenseCaptionsResult:
description: A list of captions.
required:
- values
type: object
properties:
values:
description: A list of captions.
type: array
items:
$ref: '#/definitions/DenseCaption'
Description:
description: A brief description of what the image depicts.
required:
- confidence
- text
type: object
properties:
text:
description: The text of the caption.
minLength: 1
type: string
confidence:
format: double
description: The level of confidence the service has in the caption.
maximum: 1
minimum: 0
type: number
DescriptionResult:
description: A list of descriptions sorted by confidence level.
required:
- values
type: object
properties:
values:
description: A list of descriptions sorted by confidence level.
type: array
items:
$ref: '#/definitions/Description'
DetectedObject:
description: Describes a detected object in an image.
required:
- boundingBox
- tags
type: object
properties:
id:
description: Id of the detected object.
minLength: 1
type: string
boundingBox:
$ref: '#/definitions/BoundingBox'
tags:
description: Classification confidences of the detected object.
minItems: 1
type: array
items:
$ref: '#/definitions/Tag'
DetectedPerson:
description: A person detected in an image.
required:
- boundingBox
- confidence
type: object
properties:
boundingBox:
$ref: '#/definitions/BoundingBox'
confidence:
format: double
description: >-
Confidence score of having observed the person in the image, as a
value ranging from 0 to 1.
maximum: 1
minimum: 0
type: number
DocumentLanguage:
description: An object representing the detected language for a given text span.
required:
- confidence
- languageCode
- spans
type: object
properties:
spans:
description: >-
Location of the text elements in the concatenated content the language
applies to.
type: array
items:
$ref: '#/definitions/DocumentSpan'
languageCode:
description: >-
Detected language. Value may an ISO 639-1 language code (ex. "en",
"fr") or BCP 47 language tag (ex. "zh-Hans").
minLength: 1
type: string
confidence:
format: double
description: Confidence of correctly identifying the language.
maximum: 1
minimum: 0
type: number
DocumentLine:
description: >-
A content line object consisting of an adjacent sequence of content
elements, such as words and selection marks.
required:
- boundingBox
- content
- spans
type: object
properties:
content:
description: Concatenated content of the contained elements in reading order.
minLength: 1
type: string
boundingBox:
description: Bounding box of the line.
type: array
items:
format: double
type: number
spans:
description: Location of the line in the reading order concatenated content.
type: array
items:
$ref: '#/definitions/DocumentSpan'
DocumentPage:
description: The content and layout elements extracted from a page from the input.
required:
- angle
- height
- lines
- pageNumber
- spans
- width
- words
type: object
properties:
height:
format: double
description: The height of the image/PDF in pixels/inches, respectively.
type: number
width:
format: double
description: The width of the image/PDF in pixels/inches, respectively.
type: number
angle:
format: double
description: >-
The general orientation of the content in clockwise direction,
measured in degrees between (-180, 180].
type: number
pageNumber:
format: int32
description: 1-based page number in the input document.
type: integer
words:
description: Extracted words from the page.
type: array
items:
$ref: '#/definitions/DocumentWord'
spans:
description: Location of the page in the reading order concatenated content.
type: array
items:
$ref: '#/definitions/DocumentSpan'
lines:
description: >-
Extracted lines from the page, potentially containing both textual and
visual elements.
type: array
items:
$ref: '#/definitions/DocumentLine'
DocumentSpan:
description: >-
Contiguous region of the concatenated content property, specified as an
offset and length.
required:
- length
- offset
type: object
properties:
offset:
format: int32
description: Zero-based index of the content represented by the span.
type: integer
length:
format: int32
description: Number of characters in the content represented by the span.
type: integer
DocumentStyle:
description: An object representing observed text styles.
required:
- spans
type: object
properties:
isHandwritten:
description: Is content handwritten or not.
type: boolean
spans:
description: >-
Location of the text elements in the concatenated content the style
applies to.
type: array
items:
$ref: '#/definitions/DocumentSpan'
confidence:
format: double
description: Confidence of correctly identifying the style.
type: number
DocumentWord:
description: "A word object consisting of a contiguous sequence of characters. For non-space delimited languages,\r\nsuch as Chinese, Japanese, and Korean, each character is represented as its own word."
required:
- boundingBox
- confidence
- content
- span
type: object
properties:
content:
description: Text content of the word.
minLength: 1
type: string
boundingBox:
description: Bounding box of the word.
type: array
items:
format: double
type: number
confidence:
format: double
description: Confidence of correctly extracting the word.
type: number
span:
$ref: '#/definitions/DocumentSpan'
ErrorResponse:
description: Response returned when an error occurs.
required:
- error
type: object
properties:
error:
$ref: '#/definitions/ErrorResponseDetails'
ErrorResponseDetails:
description: Error info.
required:
- code
- message
type: object
properties:
code:
description: Error code.
type: string
message:
description: Error message.
type: string
target:
description: Target of the error.
type: string
details:
description: List of detailed errors.
type: array
items:
$ref: '#/definitions/ErrorResponseDetails'
innererror:
$ref: '#/definitions/ErrorResponseInnerError'
ErrorResponseInnerError:
description: Detailed error.
required:
- code
- message
type: object
properties:
code:
description: Error code.
type: string
message:
description: Error message.
type: string
innererror:
$ref: '#/definitions/ErrorResponseInnerError'
FixtureApiModel:
description: Describes a fixture in a planogram.
required:
- h
- id
- w
- x
- 'y'
type: object
properties:
id:
description: Id of the fixture.
minLength: 1
type: string
w:
format: double
description: Width of the fixture.
type: number
h:
format: double
description: Height of the fixture.
type: number
x:
format: double
description: Left offset from the origin, in unit of in inches or centimeters.
minimum: 0
type: number
'y':
format: double
description: Top offset from the origin, in unit of in inches or centimeters.
minimum: 0
type: number
ImageAnalysisResult:
description: Describe the combined results of different types of image analysis.
required:
- metadata
- modelVersion
type: object
properties:
customModelResult:
$ref: '#/definitions/ImagePredictionResult'
captionResult:
$ref: '#/definitions/CaptionResult'
objectsResult:
$ref: '#/definitions/ObjectsResult'
readResult:
$ref: '#/definitions/ReadResult'
denseCaptionsResult:
$ref: '#/definitions/DenseCaptionsResult'
modelVersion:
description: Model Version.
minLength: 1
type: string
metadata:
$ref: '#/definitions/ImageMetadataApiModel'
tagsResult:
$ref: '#/definitions/TagsResult'
adultResult:
$ref: '#/definitions/AdultResult'
smartCropsResult:
$ref: '#/definitions/SmartCropsResult'
peopleResult:
$ref: '#/definitions/PeopleResult'
ImageMetadataApiModel:
description: The image metadata information such as height and width.
required:
- height
- width
type: object
properties:
width:
format: int32
description: The width of the image in pixels.
minimum: 1
type: integer
height:
format: int32
description: The height of the image in pixels.
minimum: 1
type: integer
ImagePredictionResult:
description: Describes the prediction result of an image.
type: object
properties:
tagsResult:
$ref: '#/definitions/TagsResult'
objectsResult:
$ref: '#/definitions/ObjectsResult'
ImageRectificationControlPointsApiModel:
description: Four corner control points for rectification. The origin is at top-left.
required:
- bottomLeft
- bottomRight
- topLeft
- topRight
type: object
properties:
topLeft:
$ref: '#/definitions/NormalizedCoordinateApiModel'
topRight:
$ref: '#/definitions/NormalizedCoordinateApiModel'
bottomLeft:
$ref: '#/definitions/NormalizedCoordinateApiModel'
bottomRight:
$ref: '#/definitions/NormalizedCoordinateApiModel'
ImageRectificationRequestApiModel:
description: Image rectification input.
required:
- controlPoints
- url
type: object
properties:
url:
format: uri
description: Source image blob URL.
type: string
controlPoints:
$ref: '#/definitions/ImageRectificationControlPointsApiModel'
ImageStitchingRequestApiModel:
description: Image stitching input.
required:
- images
type: object
properties:
images:
description: Source images to stitch.
maxItems: 20
minItems: 2
type: array
items:
format: uri
type: string
ImageUrl:
description: A JSON document with a URL pointing to the image that is to be analyzed.
required:
- url
type: object
properties:
url:
description: Publicly reachable URL of an image.
type: string
NormalizedCoordinateApiModel:
description: Normalized XY-coordinate.
required:
- x
- 'y'
type: object
properties:
x:
format: double
description: Horizontal coordinate.
maximum: 1
minimum: 0
type: number
'y':
format: double
description: Vertical coordinate.
maximum: 1
minimum: 0
type: number
ObjectsResult:
description: Describes detected objects in an image.
required:
- values
type: object
properties:
values:
description: An array of detected objects.
type: array
items:
$ref: '#/definitions/DetectedObject'
PeopleResult:
description: An object describing whether the image contains people.
required:
- values
type: object
properties:
values:
description: An array of detected people.
type: array
items:
$ref: '#/definitions/DetectedPerson'
PlanogramApiModel:
description: Describes the planogram for planogram matching operations.
required:
- fixtures
- height
- positions
- products
- width
type: object
properties:
width:
format: double
description: Width of the planogram.
type: number
height:
format: double
description: Height of the planogram.
type: number
products:
description: List of products in the planogram.
minItems: 1
type: array
items:
$ref: '#/definitions/ProductApiModel'
fixtures:
description: List of fixtures in the planogram.
minItems: 1
type: array
items:
$ref: '#/definitions/FixtureApiModel'
positions:
description: List of positions in the planogram.
minItems: 1
type: array
items:
$ref: '#/definitions/PositionApiModel'
PlanogramMatchingRequestApiModel:
description: Input to pass into the planogram matching operation.
required:
- detectedProducts
- planogram
type: object
properties:
detectedProducts:
$ref: '#/definitions/ProductRecognitionResultApiModel'
planogram:
$ref: '#/definitions/PlanogramApiModel'
PlanogramMatchingResultApiModel:
description: Results from the planogram matching operation.
required:
- matchingResultPerPosition
type: object
properties:
matchingResultPerPosition:
description: The matched detected object information for each planogram position.
type: array
items:
$ref: '#/definitions/PositionMatchingResultApiModel'
PositionApiModel:
description: Describes a product position in a planogram.
required:
- fixtureId
- id
- productId
- x
- 'y'
type: object
properties:
id:
description: Id of the position.
minLength: 1
type: string
productId:
description: Id of the product.
minLength: 1
type: string
fixtureId:
description: Id of the fixture that the product is on.
minLength: 1
type: string
x:
format: double
description: Left offset from the origin, in unit of in inches or centimeters.
minimum: 0
type: number
'y':
format: double
description: Top offset from the origin, in unit of in inches or centimeters.
minimum: 0
type: number
PositionMatchingResultApiModel:
description: >-
Paired planogram position ID and corresponding detected object from
product understanding result.
required:
- detectedObject
- positionId
type: object
properties:
positionId:
description: >-
The position ID from the planogram matched to the corresponding
detected object.
minLength: 1
type: string
detectedObject:
$ref: '#/definitions/DetectedObject'
ProductApiModel:
description: Describes a product in the planogram.
required:
- h
- id
- name
- w
type: object
properties:
id:
description: Id of the product.
minLength: 1
type: string
name:
description: Name of the product.
maxLength: 255
minLength: 1
type: string
w:
format: double
description: Width of the product.
type: number
h:
format: double
description: Height of the fixture.
type: number
ProductRecognitionApiModel:
required:
- createdDateTime
- modelName
- result
- runName
- status
- updatedDateTime
type: object
properties:
runName:
description: >-
Read only. The name that is used to uniquely identify the product
recognition run.
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
type: string
modelName:
description: Read only. The model to run product recognition against.
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
type: string
createdDateTime:
format: date-time
description: >-
Read only. The date and time when the product recognition run was
first created, in UTC.
type: string
updatedDateTime:
format: date-time
description: >-
Read only. The date and time when the product recognition run was last
updated, in UTC.
type: string
status:
description: Read only. The current state of the product recognition run.
enum:
- notStarted
- running
- succeeded
- failed
type: string
x-ms-enum:
name: ProductRecognitionState
modelAsString: true
error:
$ref: '#/definitions/ErrorResponseDetails'
result:
$ref: '#/definitions/ProductRecognitionResultApiModel'
ProductRecognitionApiModelCollectionApiModel:
description: Contains an array of results that may be paginated.
required:
- value
type: object
properties:
value:
description: The array of results.
type: array
items:
$ref: '#/definitions/ProductRecognitionApiModel'
nextLink:
description: >-
A link to the next set of paginated results, if there are more results
available; not present otherwise.
type: string
ProductRecognitionResultApiModel:
description: Results from the product understanding operation.
required:
- gaps
- imageMetadata
- products
type: object
properties:
imageMetadata:
$ref: '#/definitions/ImageMetadataApiModel'
products:
description: Products detected in the image.
type: array
items:
$ref: '#/definitions/DetectedObject'
gaps:
description: Gaps detected in the image.
type: array
items:
$ref: '#/definitions/DetectedObject'
ReadResult:
description: The results of an Read operation.
required:
- content
- pages
- stringIndexType
type: object
properties:
stringIndexType:
description: >-
The method used to compute string offset and length, possible values
include: 'textElements', 'unicodeCodePoint', 'utf16CodeUnit' etc.
minLength: 1
type: string
content:
description: >-
Concatenate string representation of all textual and visual elements
in reading order.
minLength: 1
type: string
pages:
description: A list of analyzed pages.
type: array
items:
$ref: '#/definitions/DocumentPage'
styles:
description: Extracted font styles.
type: array
items:
$ref: '#/definitions/DocumentStyle'
SingleVectorResultApiModel:
description: Results of image vectorization.
type: object
properties:
vector:
description: Vector of the image.
type: array
items:
format: float
type: number
modelVersion:
description: Model version.
type: string
SmartCropsResult:
description: Smart cropping result.
required:
- values
type: object
properties:
values:
description: Recommended regions for cropping the image.
type: array
items:
$ref: '#/definitions/CropRegion'
Tag:
description: An entity observation in the image, along with the confidence score.
required:
- confidence
- name
type: object
properties:
name:
description: Name of the entity.
minLength: 1
type: string
confidence:
format: double
description: The level of confidence that the entity was observed.
maximum: 1
minimum: 0
type: number
TagsResult:
description: A list of tags with confidence level.
required:
- values
type: object
properties:
values:
description: A list of tags with confidence level.
type: array
items:
$ref: '#/definitions/Tag'
VectorizeTextRequestApiModel:
description: Model for VectorizeText request.
required:
- text
type: object
properties:
text:
description: Text for vectorization.
minLength: 1
type: string
Dataset:
description: >-
Describes a dataset, which represents a set of images and annotations that
can be used for training or testing a model.
type: object
properties:
annotationKind:
description: "The kind of annotations contained in the annotation files.\r\nFor example, \"ImageClassification\" to specify that the annotation file contain object labels for training or testing a multiclass image classification model."
enum:
- imageClassification
- imageObjectDetection
type: string
x-ms-enum:
name: AnnotationKind
modelAsString: true
annotationFileUris:
description: "List of absolute URIs to annotation files, each of which must be stored as a blob in an Azure Storage blob container.\r\nEach file must follow the COCO format for the specified AnnotationKind, with each image path being an absolute URI to a blob in a blob container.\r\nThe Computer Vision resource must have permission to read the annotation files and all referenced image files.\r\nThis can be done by turning on System managed identities for the Computer Vision resource, then assigning the identity to a role that has permission to read from the blob container containing the annotation and image files."
maxItems: 10
type: array
items:
format: uri
type: string
customProperties:
description: >-
Dictionary of arbitrary key-value pairs for use by the application. A
maximum of 10 key-value pairs are allowed.
maxLength: 10
type: object
additionalProperties:
type: string
name:
description: Read only. The name that is used to uniquely identify the dataset.
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
type: string
createdDateTime:
format: date-time
description: >-
Read only. The date and time when the dataset was first registered, in
UTC.
type: string
updatedDateTime:
format: date-time
description: >-
Read only. The date and time when the dataset was last updated, in
UTC.
type: string
eTag:
description: >-
Read only. The latest ETag for the dataset. May be used with the
'If-Match' header when updating or deleting a dataset.
type: string
authentication:
$ref: '#/definitions/StorageAuthenticationSettingsApiModel'
DatasetApiModelCollectionApiModel:
description: Contains an array of results that may be paginated.
required:
- value
type: object
properties:
value:
description: The array of results.
type: array
items:
$ref: '#/definitions/Dataset'
nextLink:
description: >-
A link to the next set of paginated results, if there are more results
available; not present otherwise.
type: string
Model:
description: Describes a training run for training a custom model.
required:
- trainingParameters
type: object
properties:
trainingParameters:
$ref: '#/definitions/TrainingParameters'
name:
description: >-
Read only. The name that is used to uniquely identify the training
run.
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
type: string
createdDateTime:
format: date-time
description: >-
Read only. The date and time when the training run was first created,
in UTC.
type: string
updatedDateTime:
format: date-time
description: >-
Read only. The date and time when the training run was last updated,
in UTC.
type: string
status:
description: Read only. The current state of the training run.
enum:
- notStarted
- training
- succeeded
- failed
- cancelling
- cancelled
type: string
x-ms-enum:
name: ModelState
modelAsString: true
trainingCostInMinutes:
format: int32
description: >-
Read only. Actual training cost consumed, in minutes. Present only if
the training run as completed.
type: integer
readOnly: true
error:
$ref: '#/definitions/ErrorResponseDetails'
modelPerformance:
$ref: '#/definitions/ModelPerformance'
evaluationParameters:
$ref: '#/definitions/ModelEvaluationParameters'
ModelApiModelCollectionApiModel:
description: Contains an array of results that may be paginated.
required:
- value
type: object
properties:
value:
description: The array of results.
type: array
items:
$ref: '#/definitions/Model'
nextLink:
description: >-
A link to the next set of paginated results, if there are more results
available; not present otherwise.
type: string
ModelEvaluation:
description: >-
Describes an evaluation run for evaluating the accuracy of a model using a
test set.
required:
- evaluationParameters
type: object
properties:
name:
description: >-
Read only. The name that is used to uniquely identify the evaluation
run.
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
type: string
modelName:
description: Read only. The model to evaluate.
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
type: string
createdDateTime:
format: date-time
description: >-
Read only. The date and time when the evaluation run was first
created, in UTC.
type: string
updatedDateTime:
format: date-time
description: >-
Read only. The date and time when the evaluation run was last updated,
in UTC.
type: string
status:
description: Read only. The current state of the evaluation run.
enum:
- notStarted
- running
- succeeded
- failed
type: string
x-ms-enum:
name: ModelEvaluationState
modelAsString: true
evaluationParameters:
$ref: '#/definitions/ModelEvaluationParameters'
error:
$ref: '#/definitions/ErrorResponseDetails'
modelPerformance:
$ref: '#/definitions/ModelPerformance'
ModelEvaluationApiModelCollectionApiModel:
description: Contains an array of results that may be paginated.
required:
- value
type: object
properties:
value:
description: The array of results.
type: array
items:
$ref: '#/definitions/ModelEvaluation'
nextLink:
description: >-
A link to the next set of paginated results, if there are more results
available; not present otherwise.
type: string
ModelEvaluationParameters:
description: Parameters for specifying how a model is evaluated.
type: object
properties:
testDatasetName:
description: The dataset name used for testing.
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
type: string
ModelPerformance:
description: Performance metrics for a custom trained model.
required:
- tagPerformance
type: object
properties:
accuracyTop1:
format: double
description: >-
Read only. For multiclass classification models. The proportion of
test samples where the ground truth class matches the predicted class.
type: number
readOnly: true
accuracyTop5:
format: double
description: >-
Read only. For multiclass classification models. The proportion of
test samples where the ground truth class is in the top five predicted
classes.
type: number
readOnly: true
averagePrecision:
format: double
description: >-
Read only. A measure of the model performance, it summarizes the
precision and recall at different confidence thresholds.
type: number
readOnly: true
calibrationECE:
format: double
description: >-
Read only. For multiclass classification models. Expected calibration
error.
type: number
readOnly: true
meanAveragePrecision30:
format: double
description: >-
Read only. For object detection models. Mean average precision at a
threshold of 30%.
type: number
readOnly: true
meanAveragePrecision50:
format: double
description: >-
Read only. For object detection models. Mean average precision at a
threshold of 50%.
type: number
readOnly: true
meanAveragePrecision75:
format: double
description: >-
Read only. For object detection models. Mean average precision at a
threshold of 75%.
type: number
readOnly: true
tagPerformance:
description: Read only. Performance metrics for each tag recognized by the model.
type: object
additionalProperties:
$ref: '#/definitions/ModelTagPerformance'
ModelTagPerformance:
description: Performance metrics for each tag recognized by a custom trained model.
type: object
properties:
accuracy:
format: double
description: Read only. For multiclass models. Tag accuracy.
type: number
readOnly: true
averagePrecision50:
format: double
description: >-
Read only. For object detection models. Average precision at a
threshold of 50%.
type: number
readOnly: true
StorageAuthenticationSettingsApiModel:
description: Describes the storage authentication settings.
required:
- kind
type: object
properties:
kind:
description: The storage authentication kind.
enum:
- none
- managedIdentity
- sas
type: string
x-ms-enum:
name: StorageAuthenticationKind
modelAsString: true
sasToken:
description: >-
Optional. The sas token to access container. Only needed when Kind =
Sas.
type: string
TrainingParameters:
description: Parameters for specifying how a training run trains a custom model.
required:
- timeBudgetInHours
- trainingDatasetName
type: object
properties:
trainingDatasetName:
description: The dataset name used for training.
maxLength: 255
pattern: ^[a-zA-Z0-9][a-zA-Z0-9._-]*$
type: string
timeBudgetInHours:
format: int32
description: "Time budget for training, in hours. The minimum allowed value is 1, and the maximum allowed value is 336 hours for GenericClassifier, 1344 hours for GenericDetector.\r\nThis is the maximum amount of compute time that will be spent to train the model."
maximum: 1344
minimum: 1
type: integer
modelKind:
description: Model kind.
enum:
- Product-Recognizer
- Generic-Classifier
- Generic-Detector
type: string
x-ms-enum:
name: ModelKind
modelAsString: true
parameters:
ApiVersion:
in: query
name: api-version
description: Requested API version.
required: true
type: string
x-ms-parameter-location: client
x-ms-paths:
/imageanalysis:segment?overload=stream:
post:
tags:
- Operations
summary: >-
Analyze the input image. The request either contains an image stream
with any content type ['image/*', 'application/octet-stream'], or a JSON
payload which includes a url property to be used to retrieve the image
stream. An image stream of content type 'image/png' is returned, where
the pixel values depend on the analysis mode. The returned image has the
same dimensions as the input image for modes: foregroundMatting. The
returned image has the same aspect ratio and same dimensions as the
input image up to a limit of 16 megapixels for modes: backgroundRemoval.
operationId: ImageAnalysis_SegmentFromImageStream
consumes:
- application/octet-stream
- image/jpeg
- image/gif
- image/tiff
- image/bmp
- image/png
produces:
- image/png
- application/json
parameters:
- in: query
name: mode
description: The analysis mode requested.
type: string
enum:
- backgroundRemoval
- foregroundMatting
x-ms-enum:
name: SegmentationMode
modelAsString: true
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: An image stream.
required: true
schema:
format: byte
type: string
responses:
'200':
description: Success
schema:
type: file
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
Segment_BackgroundRemoval_FromImageStream:
$ref: ./examples/Segment_BackgroundRemoval_FromImageStream.json
Segment_ForegroundMatting_FromImageStream:
$ref: ./examples/Segment_ForegroundMatting_FromImageStream.json
/retrieval:vectorizeImage?overload=stream:
post:
tags:
- ImageRetrieval
summary: Return vector from an image.
operationId: ImageRetrieval_VectorizeImageFromImageStream
consumes:
- application/octet-stream
- image/jpeg
- image/gif
- image/tiff
- image/bmp
- image/png
produces:
- application/json
parameters:
- in: query
name: model-version
description: Model version.
type: string
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: An image stream.
required: true
schema:
format: byte
type: string
responses:
'200':
description: Success
schema:
$ref: '#/definitions/SingleVectorResultApiModel'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
ImageRetrieval_VectorizeImageFromImageStream:
$ref: ./examples/ImageRetrieval_VectorizeImageFromImageStream.json
/imageanalysis:analyze?overload=stream:
post:
tags:
- Operations
summary: >-
Analyze the input image. The request either contains image stream with
any content type ['image/*', 'application/octet-stream'], or a JSON
payload which includes an url property to be used to retrieve the image
stream.
operationId: ImageAnalysis_AnalyzeFromImageStream
consumes:
- application/octet-stream
- image/jpeg
- image/gif
- image/tiff
- image/bmp
- image/png
produces:
- application/json
parameters:
- in: query
name: features
description: >-
The visual features requested: tags, objects, caption,
denseCaptions, read, smartCrops, people. This parameter needs to be
specified if the parameter "model-name" is not specified.
type: array
items:
enum:
- tags
- caption
- denseCaptions
- objects
- read
- smartCrops
- people
type: string
x-ms-enum:
name: VisualFeature
modelAsString: true
collectionFormat: csv
- in: query
name: model-name
description: >-
The name of the custom trained model. This parameter needs to be
specified if the parameter "features" is not specified.
type: string
- in: query
name: language
description: >-
The desired language for output generation. If this parameter is not
specified, the default value is "en". See
https://aka.ms/cv-languages for a list of supported languages.
type: string
default: en
- in: query
name: smartcrops-aspect-ratios
description: >-
A list of aspect ratios to use for smartCrops feature. Aspect ratios
are calculated by dividing the target crop width by the height.
Supported values are between 0.75 and 1.8 (inclusive). Multiple
values should be comma-separated. If this parameter is not
specified, the service will return one crop suggestion with an
aspect ratio it sees fit between 0.5 and 2.0 (inclusive).
type: string
- in: query
name: gender-neutral-caption
description: >-
Boolean flag for enabling gender-neutral captioning for caption and
denseCaptions features. If this parameter is not specified, the
default value is "false".
type: boolean
default: false
- $ref: '#/parameters/ApiVersion'
- in: body
name: body
description: An image stream.
required: true
schema:
format: byte
type: string
responses:
'200':
description: Success
schema:
$ref: '#/definitions/ImageAnalysisResult'
default:
description: Error
schema:
$ref: '#/definitions/ErrorResponse'
headers:
x-ms-error-code:
type: string
x-ms-examples:
AnalyzeImageFromImageStream_CustomModel:
$ref: ./examples/AnalyzeImageFromImageStream_CustomModel.json
tags:
- name: Datasets
- name: ImageRetrieval
- name: Models
- name: Operations