IBatchDeployment Interface

ARM resource ID of the code asset.

[Required] The script to execute on startup. eg. "score.py"

Compute target for batch inference operation.

[Required] The type of the deployment

Description of the endpoint deployment.

Property dictionary. Properties can be added, but not removed or altered.

ARM resource ID or AssetId of the environment specification for the endpoint deployment.

Environment variables configuration for the deployment.

Error threshold, if the error count for the entire input goes above this value, the batch inference will be aborted. Range is [-1, int.MaxValue]. For FileDataset, this value is the count of file failures. For TabularDataset, this value is the count of record failures. If set to -1 (the lower bound), all failures during batch inference will be ignored.

Fully qualified resource ID for the resource. Ex - /subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/{resourceProviderNamespace}/{resourceType}/{resourceName}

The service principal ID of the system assigned identity. This property will only be provided for a system assigned identity.

The tenant ID of the system assigned identity. This property will only be provided for a system assigned identity.

Type of managed service identity (where both SystemAssigned and UserAssigned types are allowed).

The set of user assigned identities associated with the resource. The userAssignedIdentities dictionary keys will be ARM resource ids in the form: '/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.ManagedIdentity/userAssignedIdentities/{identityName}. The dictionary values can be empty objects ({}) in requests.

Metadata used by portal/tooling/etc to render different UX experiences for resources of the same type.

The geo-location where the resource lives

Logging level for batch inference operation.

Indicates maximum number of parallelism per instance.

Size of the mini-batch passed to each batch invocation. For FileDataset, this is the number of files per mini-batch. For TabularDataset, this is the size of the records in bytes, per mini-batch.

Reference to the model asset for the endpoint deployment.

The name of the resource

Indicates how the output will be organized.

Customized output file name for append_row output action.

Provisioning state for the endpoint deployment.

Optional number of instances or nodes used by the compute target.

Optional type of VM used as supported by the compute target.

Additional properties bag.

Maximum retry count for a mini-batch

Invocation timeout for a mini-batch, in ISO 8601 format.

If the SKU supports scale out/in then the capacity integer should be included. If scale out/in is not possible for the resource this may be omitted.

If the service has different generations of hardware, for the same SKU, then that can be captured here.

The name of the SKU. Ex - P3. It is typically a letter+number code

The SKU size. When the name field is the combination of tier and some other value, this would be the standalone code.

This field is required to be implemented by the Resource Provider if the service has more than one tier, but is not required on a PUT.

The timestamp of resource creation (UTC).

The identity that created the resource.

The type of identity that created the resource.

The timestamp of resource last modification (UTC)

The identity that last modified the resource.

The type of identity that last modified the resource.

Resource tags.

The type of the resource. E.g. "Microsoft.Compute/virtualMachines" or "Microsoft.Storage/storageAccounts"