Microsoft.MachineLearning webServices

Article
02/13/2023

Bicep resource definition

The webServices resource type can be deployed to:

Resource groups - See resource group deployment commands

For a list of changed properties in each API version, see change log.

Resource format

To create a Microsoft.MachineLearning/webServices resource, add the following Bicep to your template.

resource symbolicname 'Microsoft.MachineLearning/webServices@2017-01-01' = {
  name: 'string'
  location: 'string'
  tags: {
    tagName1: 'tagValue1'
    tagName2: 'tagValue2'
  }
  properties: {
    assets: {}
    commitmentPlan: {
      id: 'string'
    }
    description: 'string'
    diagnostics: {
      expiry: 'string'
      level: 'string'
    }
    exampleRequest: {
      globalParameters: {}
      inputs: {}
    }
    exposeSampleData: bool
    input: {
      description: 'string'
      properties: {}
      title: 'string'
      type: 'string'
    }
    keys: {
      primary: 'string'
      secondary: 'string'
    }
    machineLearningWorkspace: {
      id: 'string'
    }
    output: {
      description: 'string'
      properties: {}
      title: 'string'
      type: 'string'
    }
    parameters: {}
    payloadsInBlobStorage: bool
    payloadsLocation: {
      credentials: 'string'
      uri: 'string'
    }
    readOnly: bool
    realtimeConfiguration: {
      maxConcurrentCalls: int
    }
    storageAccount: {
      key: 'string'
      name: 'string'
    }
    title: 'string'
    packageType: 'string'
    // For remaining properties, see WebServiceProperties objects
  }
}

WebServiceProperties objects

Set the packageType property to specify the type of object.

For Graph, use:

  packageType: 'Graph'
  package: {
    edges: [
      {
        sourceNodeId: 'string'
        sourcePortId: 'string'
        targetNodeId: 'string'
        targetPortId: 'string'
      }
    ]
    graphParameters: {}
    nodes: {}
  }

Property values

webServices

Name	Description	Value
name	The resource name	string (required) Character limit: 1-260 Valid characters: Can't use: `<>*%&:?+/\\` or control characters Can't end with a space.
location	Specifies the location of the resource.	string (required)
tags	Contains resource tags defined as key/value pairs.	Dictionary of tag names and values. See Tags in templates
properties	Contains the property payload that describes the web service.	WebServiceProperties (required)

WebServiceProperties

Name	Description	Value
assets	Contains user defined properties describing web service assets. Properties are expressed as Key/Value pairs.	object
commitmentPlan	Contains the commitment plan associated with this web service. Set at creation time. Once set, this value cannot be changed. Note: The commitment plan is not returned from calls to GET operations.	CommitmentPlan
description	The description of the web service.	string
diagnostics	Settings controlling the diagnostics traces collection for the web service.	DiagnosticsConfiguration
exampleRequest	Defines sample input data for one or more of the service's inputs.	ExampleRequest
exposeSampleData	When set to true, sample data is included in the web service's swagger definition. The default value is true.	bool
input	Contains the Swagger 2.0 schema describing one or more of the web service's inputs. For more information, see the Swagger specification.	ServiceInputOutputSpecification
keys	Contains the web service provisioning keys. If you do not specify provisioning keys, the Azure Machine Learning system generates them for you. Note: The keys are not returned from calls to GET operations.	WebServiceKeys
machineLearningWorkspace	Specifies the Machine Learning workspace containing the experiment that is source for the web service.	MachineLearningWorkspace
output	Contains the Swagger 2.0 schema describing one or more of the web service's outputs. For more information, see the Swagger specification.	ServiceInputOutputSpecification
parameters	The set of global parameters values defined for the web service, given as a global parameter name to default value map. If no default value is specified, the parameter is considered to be required.	object
payloadsInBlobStorage	When set to true, indicates that the payload size is larger than 3 MB. Otherwise false. If the payload size exceed 3 MB, the payload is stored in a blob and the PayloadsLocation parameter contains the URI of the blob. Otherwise, this will be set to false and Assets, Input, Output, Package, Parameters, ExampleRequest are inline. The Payload sizes is determined by adding the size of the Assets, Input, Output, Package, Parameters, and the ExampleRequest.	bool
payloadsLocation	The URI of the payload blob. This parameter contains a value only if the payloadsInBlobStorage parameter is set to true. Otherwise is set to null.	BlobLocation
readOnly	When set to true, indicates that the web service is read-only and can no longer be updated or patched, only removed. Default, is false. Note: Once set to true, you cannot change its value.	bool
realtimeConfiguration	Contains the configuration settings for the web service endpoint.	RealtimeConfiguration
storageAccount	Specifies the storage account that Azure Machine Learning uses to store information about the web service. Only the name of the storage account is returned from calls to GET operations. When updating the storage account information, you must ensure that all necessary assets are available in the new storage account or calls to your web service will fail.	StorageAccount
title	The title of the web service.	string
packageType	Set the object type	Graph (required)

CommitmentPlan

Name	Description	Value
id	Specifies the Azure Resource Manager ID of the commitment plan associated with the web service.	string (required)

DiagnosticsConfiguration

Name	Description	Value
expiry	Specifies the date and time when the logging will cease. If null, diagnostic collection is not time limited.	string
level	Specifies the verbosity of the diagnostic output. Valid values are: None - disables tracing; Error - collects only error (stderr) traces; All - collects all traces (stdout and stderr).	'All' 'Error' 'None' (required)

ExampleRequest

Name	Description	Value
globalParameters	Sample input data for the web service's global parameters	object
inputs	Sample input data for the web service's input(s) given as an input name to sample input values matrix map.	object

ServiceInputOutputSpecification

Name	Description	Value
description	The description of the Swagger schema.	string
properties	Specifies a collection that contains the column schema for each input or output of the web service. For more information, see the Swagger specification.	object (required)
title	The title of your Swagger schema.	string
type	The type of the entity described in swagger. Always 'object'.	string (required)

WebServiceKeys

Name	Description	Value
primary	The primary access key.	string
secondary	The secondary access key.	string

MachineLearningWorkspace

Name	Description	Value
id	Specifies the workspace ID of the machine learning workspace associated with the web service	string (required)

BlobLocation

Name	Description	Value
credentials	Access credentials for the blob, if applicable (e.g. blob specified by storage account connection string + blob URI)	string
uri	The URI from which the blob is accessible from. For example, aml://abc for system assets or `https://xyz` for user assets or payload.	string (required)

RealtimeConfiguration

Name	Description	Value
maxConcurrentCalls	Specifies the maximum concurrent calls that can be made to the web service. Minimum value: 4, Maximum value: 200.	int

StorageAccount

Name	Description	Value
key	Specifies the key used to access the storage account.	string
name	Specifies the name of the storage account.	string

WebServicePropertiesForGraph

Name	Description	Value
packageType	Specifies the package type. Valid values are Graph (Specifies a web service published through the Machine Learning Studio) and Code (Specifies a web service published using code such as Python). Note: Code is not supported at this time.	'Graph' (required)
package	The definition of the graph package making up this web service.	GraphPackage

GraphPackage

Name	Description	Value
edges	The list of edges making up the graph.	GraphEdge[]
graphParameters	The collection of global parameters for the graph, given as a global parameter name to GraphParameter map. Each parameter here has a 1:1 match with the global parameters values map declared at the WebServiceProperties level.	object
nodes	The set of nodes making up the graph, provided as a nodeId to GraphNode map	object

GraphEdge

Name	Description	Value
sourceNodeId	The source graph node's identifier.	string
sourcePortId	The identifier of the source node's port that the edge connects from.	string
targetNodeId	The destination graph node's identifier.	string
targetPortId	The identifier of the destination node's port that the edge connects into.	string

ARM template resource definition

The webServices resource type can be deployed to:

Resource groups - See resource group deployment commands

For a list of changed properties in each API version, see change log.

Resource format

To create a Microsoft.MachineLearning/webServices resource, add the following JSON to your template.

{
  "type": "Microsoft.MachineLearning/webServices",
  "apiVersion": "2017-01-01",
  "name": "string",
  "location": "string",
  "tags": {
    "tagName1": "tagValue1",
    "tagName2": "tagValue2"
  },
  "properties": {
    "assets": {},
    "commitmentPlan": {
      "id": "string"
    },
    "description": "string",
    "diagnostics": {
      "expiry": "string",
      "level": "string"
    },
    "exampleRequest": {
      "globalParameters": {},
      "inputs": {}
    },
    "exposeSampleData": "bool",
    "input": {
      "description": "string",
      "properties": {},
      "title": "string",
      "type": "string"
    },
    "keys": {
      "primary": "string",
      "secondary": "string"
    },
    "machineLearningWorkspace": {
      "id": "string"
    },
    "output": {
      "description": "string",
      "properties": {},
      "title": "string",
      "type": "string"
    },
    "parameters": {},
    "payloadsInBlobStorage": "bool",
    "payloadsLocation": {
      "credentials": "string",
      "uri": "string"
    },
    "readOnly": "bool",
    "realtimeConfiguration": {
      "maxConcurrentCalls": "int"
    },
    "storageAccount": {
      "key": "string",
      "name": "string"
    },
    "title": "string",
    "packageType": "string"
    // For remaining properties, see WebServiceProperties objects
  }
}

WebServiceProperties objects

Set the packageType property to specify the type of object.

For Graph, use:

  "packageType": "Graph",
  "package": {
    "edges": [
      {
        "sourceNodeId": "string",
        "sourcePortId": "string",
        "targetNodeId": "string",
        "targetPortId": "string"
      }
    ],
    "graphParameters": {},
    "nodes": {}
  }

Property values

webServices

Name	Description	Value
type	The resource type	'Microsoft.MachineLearning/webServices'
apiVersion	The resource api version	'2017-01-01'
name	The resource name	string (required) Character limit: 1-260 Valid characters: Can't use: `<>*%&:?+/\\` or control characters Can't end with a space.
location	Specifies the location of the resource.	string (required)
tags	Contains resource tags defined as key/value pairs.	Dictionary of tag names and values. See Tags in templates
properties	Contains the property payload that describes the web service.	WebServiceProperties (required)

WebServiceProperties

Name	Description	Value
assets	Contains user defined properties describing web service assets. Properties are expressed as Key/Value pairs.	object
commitmentPlan	Contains the commitment plan associated with this web service. Set at creation time. Once set, this value cannot be changed. Note: The commitment plan is not returned from calls to GET operations.	CommitmentPlan
description	The description of the web service.	string
diagnostics	Settings controlling the diagnostics traces collection for the web service.	DiagnosticsConfiguration
exampleRequest	Defines sample input data for one or more of the service's inputs.	ExampleRequest
exposeSampleData	When set to true, sample data is included in the web service's swagger definition. The default value is true.	bool
input	Contains the Swagger 2.0 schema describing one or more of the web service's inputs. For more information, see the Swagger specification.	ServiceInputOutputSpecification
keys	Contains the web service provisioning keys. If you do not specify provisioning keys, the Azure Machine Learning system generates them for you. Note: The keys are not returned from calls to GET operations.	WebServiceKeys
machineLearningWorkspace	Specifies the Machine Learning workspace containing the experiment that is source for the web service.	MachineLearningWorkspace
output	Contains the Swagger 2.0 schema describing one or more of the web service's outputs. For more information, see the Swagger specification.	ServiceInputOutputSpecification
parameters	The set of global parameters values defined for the web service, given as a global parameter name to default value map. If no default value is specified, the parameter is considered to be required.	object
payloadsInBlobStorage	When set to true, indicates that the payload size is larger than 3 MB. Otherwise false. If the payload size exceed 3 MB, the payload is stored in a blob and the PayloadsLocation parameter contains the URI of the blob. Otherwise, this will be set to false and Assets, Input, Output, Package, Parameters, ExampleRequest are inline. The Payload sizes is determined by adding the size of the Assets, Input, Output, Package, Parameters, and the ExampleRequest.	bool
payloadsLocation	The URI of the payload blob. This parameter contains a value only if the payloadsInBlobStorage parameter is set to true. Otherwise is set to null.	BlobLocation
readOnly	When set to true, indicates that the web service is read-only and can no longer be updated or patched, only removed. Default, is false. Note: Once set to true, you cannot change its value.	bool
realtimeConfiguration	Contains the configuration settings for the web service endpoint.	RealtimeConfiguration
storageAccount	Specifies the storage account that Azure Machine Learning uses to store information about the web service. Only the name of the storage account is returned from calls to GET operations. When updating the storage account information, you must ensure that all necessary assets are available in the new storage account or calls to your web service will fail.	StorageAccount
title	The title of the web service.	string
packageType	Set the object type	Graph (required)

CommitmentPlan

Name	Description	Value
id	Specifies the Azure Resource Manager ID of the commitment plan associated with the web service.	string (required)

DiagnosticsConfiguration

Name	Description	Value
expiry	Specifies the date and time when the logging will cease. If null, diagnostic collection is not time limited.	string
level	Specifies the verbosity of the diagnostic output. Valid values are: None - disables tracing; Error - collects only error (stderr) traces; All - collects all traces (stdout and stderr).	'All' 'Error' 'None' (required)

ExampleRequest

Name	Description	Value
globalParameters	Sample input data for the web service's global parameters	object
inputs	Sample input data for the web service's input(s) given as an input name to sample input values matrix map.	object

ServiceInputOutputSpecification

Name	Description	Value
description	The description of the Swagger schema.	string
properties	Specifies a collection that contains the column schema for each input or output of the web service. For more information, see the Swagger specification.	object (required)
title	The title of your Swagger schema.	string
type	The type of the entity described in swagger. Always 'object'.	string (required)

WebServiceKeys

Name	Description	Value
primary	The primary access key.	string
secondary	The secondary access key.	string

MachineLearningWorkspace

Name	Description	Value
id	Specifies the workspace ID of the machine learning workspace associated with the web service	string (required)

BlobLocation

Name	Description	Value
credentials	Access credentials for the blob, if applicable (e.g. blob specified by storage account connection string + blob URI)	string
uri	The URI from which the blob is accessible from. For example, aml://abc for system assets or `https://xyz` for user assets or payload.	string (required)

RealtimeConfiguration

Name	Description	Value
maxConcurrentCalls	Specifies the maximum concurrent calls that can be made to the web service. Minimum value: 4, Maximum value: 200.	int

StorageAccount

Name	Description	Value
key	Specifies the key used to access the storage account.	string
name	Specifies the name of the storage account.	string

WebServicePropertiesForGraph

Name	Description	Value
packageType	Specifies the package type. Valid values are Graph (Specifies a web service published through the Machine Learning Studio) and Code (Specifies a web service published using code such as Python). Note: Code is not supported at this time.	'Graph' (required)
package	The definition of the graph package making up this web service.	GraphPackage

GraphPackage

Name	Description	Value
edges	The list of edges making up the graph.	GraphEdge[]
graphParameters	The collection of global parameters for the graph, given as a global parameter name to GraphParameter map. Each parameter here has a 1:1 match with the global parameters values map declared at the WebServiceProperties level.	object
nodes	The set of nodes making up the graph, provided as a nodeId to GraphNode map	object

GraphEdge

Name	Description	Value
sourceNodeId	The source graph node's identifier.	string
sourcePortId	The identifier of the source node's port that the edge connects from.	string
targetNodeId	The destination graph node's identifier.	string
targetPortId	The identifier of the destination node's port that the edge connects into.	string

Terraform (AzAPI provider) resource definition

The webServices resource type can be deployed to:

Resource groups

For a list of changed properties in each API version, see change log.

Resource format

To create a Microsoft.MachineLearning/webServices resource, add the following Terraform to your template.

resource "azapi_resource" "symbolicname" {
  type = "Microsoft.MachineLearning/webServices@2017-01-01"
  name = "string"
  location = "string"
  parent_id = "string"
  tags = {
    tagName1 = "tagValue1"
    tagName2 = "tagValue2"
  }
  body = jsonencode({
    properties = {
      assets = {}
      commitmentPlan = {
        id = "string"
      }
      description = "string"
      diagnostics = {
        expiry = "string"
        level = "string"
      }
      exampleRequest = {
        globalParameters = {}
        inputs = {}
      }
      exposeSampleData = bool
      input = {
        description = "string"
        properties = {}
        title = "string"
        type = "string"
      }
      keys = {
        primary = "string"
        secondary = "string"
      }
      machineLearningWorkspace = {
        id = "string"
      }
      output = {
        description = "string"
        properties = {}
        title = "string"
        type = "string"
      }
      parameters = {}
      payloadsInBlobStorage = bool
      payloadsLocation = {
        credentials = "string"
        uri = "string"
      }
      readOnly = bool
      realtimeConfiguration = {
        maxConcurrentCalls = int
      }
      storageAccount = {
        key = "string"
        name = "string"
      }
      title = "string"
      packageType = "string"
      // For remaining properties, see WebServiceProperties objects
    }
  })
}

WebServiceProperties objects

Set the packageType property to specify the type of object.

For Graph, use:

  packageType = "Graph"
  package = {
    edges = [
      {
        sourceNodeId = "string"
        sourcePortId = "string"
        targetNodeId = "string"
        targetPortId = "string"
      }
    ]
    graphParameters = {}
    nodes = {}
  }

Property values

webServices

Name	Description	Value
type	The resource type	"Microsoft.MachineLearning/webServices@2017-01-01"
name	The resource name	string (required) Character limit: 1-260 Valid characters: Can't use: `<>*%&:?+/\\` or control characters Can't end with a space.
location	Specifies the location of the resource.	string (required)
parent_id	To deploy to a resource group, use the ID of that resource group.	string (required)
tags	Contains resource tags defined as key/value pairs.	Dictionary of tag names and values.
properties	Contains the property payload that describes the web service.	WebServiceProperties (required)

WebServiceProperties

Name	Description	Value
assets	Contains user defined properties describing web service assets. Properties are expressed as Key/Value pairs.	object
commitmentPlan	Contains the commitment plan associated with this web service. Set at creation time. Once set, this value cannot be changed. Note: The commitment plan is not returned from calls to GET operations.	CommitmentPlan
description	The description of the web service.	string
diagnostics	Settings controlling the diagnostics traces collection for the web service.	DiagnosticsConfiguration
exampleRequest	Defines sample input data for one or more of the service's inputs.	ExampleRequest
exposeSampleData	When set to true, sample data is included in the web service's swagger definition. The default value is true.	bool
input	Contains the Swagger 2.0 schema describing one or more of the web service's inputs. For more information, see the Swagger specification.	ServiceInputOutputSpecification
keys	Contains the web service provisioning keys. If you do not specify provisioning keys, the Azure Machine Learning system generates them for you. Note: The keys are not returned from calls to GET operations.	WebServiceKeys
machineLearningWorkspace	Specifies the Machine Learning workspace containing the experiment that is source for the web service.	MachineLearningWorkspace
output	Contains the Swagger 2.0 schema describing one or more of the web service's outputs. For more information, see the Swagger specification.	ServiceInputOutputSpecification
parameters	The set of global parameters values defined for the web service, given as a global parameter name to default value map. If no default value is specified, the parameter is considered to be required.	object
payloadsInBlobStorage	When set to true, indicates that the payload size is larger than 3 MB. Otherwise false. If the payload size exceed 3 MB, the payload is stored in a blob and the PayloadsLocation parameter contains the URI of the blob. Otherwise, this will be set to false and Assets, Input, Output, Package, Parameters, ExampleRequest are inline. The Payload sizes is determined by adding the size of the Assets, Input, Output, Package, Parameters, and the ExampleRequest.	bool
payloadsLocation	The URI of the payload blob. This parameter contains a value only if the payloadsInBlobStorage parameter is set to true. Otherwise is set to null.	BlobLocation
readOnly	When set to true, indicates that the web service is read-only and can no longer be updated or patched, only removed. Default, is false. Note: Once set to true, you cannot change its value.	bool
realtimeConfiguration	Contains the configuration settings for the web service endpoint.	RealtimeConfiguration
storageAccount	Specifies the storage account that Azure Machine Learning uses to store information about the web service. Only the name of the storage account is returned from calls to GET operations. When updating the storage account information, you must ensure that all necessary assets are available in the new storage account or calls to your web service will fail.	StorageAccount
title	The title of the web service.	string
packageType	Set the object type	Graph (required)

CommitmentPlan

Name	Description	Value
id	Specifies the Azure Resource Manager ID of the commitment plan associated with the web service.	string (required)

DiagnosticsConfiguration

Name	Description	Value
expiry	Specifies the date and time when the logging will cease. If null, diagnostic collection is not time limited.	string
level	Specifies the verbosity of the diagnostic output. Valid values are: None - disables tracing; Error - collects only error (stderr) traces; All - collects all traces (stdout and stderr).	"All" "Error" "None" (required)

ExampleRequest

Name	Description	Value
globalParameters	Sample input data for the web service's global parameters	object
inputs	Sample input data for the web service's input(s) given as an input name to sample input values matrix map.	object

ServiceInputOutputSpecification

Name	Description	Value
description	The description of the Swagger schema.	string
properties	Specifies a collection that contains the column schema for each input or output of the web service. For more information, see the Swagger specification.	object (required)
title	The title of your Swagger schema.	string
type	The type of the entity described in swagger. Always 'object'.	string (required)

WebServiceKeys

Name	Description	Value
primary	The primary access key.	string
secondary	The secondary access key.	string

MachineLearningWorkspace

Name	Description	Value
id	Specifies the workspace ID of the machine learning workspace associated with the web service	string (required)

BlobLocation

Name	Description	Value
credentials	Access credentials for the blob, if applicable (e.g. blob specified by storage account connection string + blob URI)	string
uri	The URI from which the blob is accessible from. For example, aml://abc for system assets or `https://xyz` for user assets or payload.	string (required)

RealtimeConfiguration

Name	Description	Value
maxConcurrentCalls	Specifies the maximum concurrent calls that can be made to the web service. Minimum value: 4, Maximum value: 200.	int

StorageAccount

Name	Description	Value
key	Specifies the key used to access the storage account.	string
name	Specifies the name of the storage account.	string

WebServicePropertiesForGraph

Name	Description	Value
packageType	Specifies the package type. Valid values are Graph (Specifies a web service published through the Machine Learning Studio) and Code (Specifies a web service published using code such as Python). Note: Code is not supported at this time.	"Graph" (required)
package	The definition of the graph package making up this web service.	GraphPackage

GraphPackage

Name	Description	Value
edges	The list of edges making up the graph.	GraphEdge[]
graphParameters	The collection of global parameters for the graph, given as a global parameter name to GraphParameter map. Each parameter here has a 1:1 match with the global parameters values map declared at the WebServiceProperties level.	object
nodes	The set of nodes making up the graph, provided as a nodeId to GraphNode map	object

GraphEdge

Name	Description	Value
sourceNodeId	The source graph node's identifier.	string
sourcePortId	The identifier of the source node's port that the edge connects from.	string
targetNodeId	The destination graph node's identifier.	string
targetPortId	The identifier of the destination node's port that the edge connects into.	string

Microsoft.MachineLearning webServices

Bicep resource definition

Resource format

WebServiceProperties objects

Property values

webServices

WebServiceProperties

CommitmentPlan

DiagnosticsConfiguration

ExampleRequest

ServiceInputOutputSpecification

WebServiceKeys

MachineLearningWorkspace

BlobLocation

RealtimeConfiguration

StorageAccount

WebServicePropertiesForGraph

GraphPackage

GraphEdge

ARM template resource definition

Resource format

WebServiceProperties objects

Property values

webServices

WebServiceProperties

CommitmentPlan

DiagnosticsConfiguration

ExampleRequest

ServiceInputOutputSpecification

WebServiceKeys

MachineLearningWorkspace

BlobLocation

RealtimeConfiguration

StorageAccount

WebServicePropertiesForGraph

GraphPackage

GraphEdge

Terraform (AzAPI provider) resource definition

Resource format

WebServiceProperties objects

Property values

webServices

WebServiceProperties

CommitmentPlan

DiagnosticsConfiguration

ExampleRequest

ServiceInputOutputSpecification

WebServiceKeys

MachineLearningWorkspace

BlobLocation

RealtimeConfiguration

StorageAccount

WebServicePropertiesForGraph

GraphPackage

GraphEdge

Feedback

Feedback

Additional resources