[ aws . elementalinference ]
Creates a function. A function analyzes a feed by building a request from the feed’s media and metadata, optionally sending that request to an Amazon Bedrock model in your account, and publishing the result to the feed’s metadata timeline.
A function runs only after you attach it to a feed as a function output. For more information, see Using functions in the AWS Elemental Inference User Guide .
See also: AWS API Documentation
create-function
--name <value>
--target <value>
--request-expression <value>
[--response-expression <value>]
[--description <value>]
[--event-bridge-delivery <value>]
[--tags <value>]
[--cli-input-json | --cli-input-yaml]
[--generate-cli-skeleton <value>]
[--debug]
[--endpoint-url <value>]
[--no-verify-ssl]
[--no-paginate]
[--output <value>]
[--query <value>]
[--profile <value>]
[--region <value>]
[--version <value>]
[--color <value>]
[--no-sign-request]
[--ca-bundle <value>]
[--cli-read-timeout <value>]
[--cli-connect-timeout <value>]
[--cli-binary-format <value>]
[--no-cli-pager]
[--cli-auto-prompt]
[--no-cli-auto-prompt]
[--cli-error-format <value>]
[--help]
--name (string) [required]
A name for this function.
Constraints:
- pattern:
[a-zA-Z0-9]([a-zA-Z0-9-_]{0,126}[a-zA-Z0-9])?
--target (tagged union structure) [required]
The destination for the request that the function builds.
Note
This is a Tagged Union structure. Only one of the following top level keys can be set:bedrock,passthrough.bedrock -> (structure)
Send the request to an Amazon Bedrock model.
api -> (string) [required]
The Amazon Bedrock API to call. The only valid value is INVOKE_MODEL.
Possible values:
INVOKE_MODELmodel -> (string) [required]
The model to invoke. Specify one of the following:
- A model ID.
- An inference profile ID.
- The ARN of a custom model.
- The ARN of a provisioned model.
Elemental Inference invokes the model in the Region of the feed. To use a model in other Regions, specify a cross-Region inference profile ID. A custom or provisioned model must be in the Region of the feed.
The request expression must produce a request body that this model accepts, and the model must accept the kind of media that the request expression references. For example, a text-only model can’t accept a request that references media.frames.
Constraints:
- min:
1- max:
2048- pattern:
([a-z0-9-]{1,63}[.][a-z0-9-.:]{1,76}|arn:aws:bedrock:[a-z0-9-]{1,20}:[0-9]{12}:(custom-model|provisioned-model)/.+)passthrough -> (structure)
Send the request nowhere. Use this target to reshape the metadata that Elemental Inference already produces, with no model invocation and no model charges.
Shorthand Syntax:
bedrock={api=string,model=string},passthrough={}
JSON Syntax:
{
"bedrock": {
"api": "INVOKE_MODEL",
"model": "string"
},
"passthrough": {
}
}
--request-expression (string) [required]
The JSONata expression that builds the request. Elemental Inference evaluates it against the media object of the feed, which holds the media and metadata that the function asked for. Maximum length: 5,120 characters.
When you save the function, Elemental Inference evaluates the expression against sample media and rejects it if the evaluation fails. For a Bedrock target, the result must be a JSON object; Elemental Inference doesn’t check that the object is a valid request body for the model. The expression can reference at most one of media.transcript, media.contextual, and media.clip. For a passthrough target, it can’t reference media.frames.
Constraints:
- min:
1- max:
5120
--response-expression (string)
The JSONata expression that selects what to store. Elemental Inference evaluates it against the result as its root: the model response for a Bedrock target, or the output of the request expression for a passthrough target. If a function has no response expression, Elemental Inference stores the entire result. Maximum length: 5,120 characters.
When you save the function, Elemental Inference checks that the expression is valid JSONata. For a passthrough target, it also evaluates the request and response expressions together against sample media, and rejects them if they produce no value.
Constraints:
- min:
0- max:
5120
--description (string)
The description of the function. Maximum length: 1,024 characters.
Constraints:
- min:
0- max:
1024- pattern:
[\w \-\.',@:;]*
--event-bridge-delivery (string)
Specifies whether Elemental Inference also sends this function’s results to Amazon EventBridge. Elemental Inference sends a result only when the result is a JSON object with a top-level
eventBridgefield set totrue. It sends the result, unchanged, as a Function Metadata Generated event to the default event bus in your account, after the result is available from GetMetadata. If the result is larger than 64 KB, the event hasresultOmittedset totruein place of the result.Valid values:
- ENABLED – Elemental Inference sends each marked result as an event.
- DISABLED (default) – Elemental Inference doesn’t send events.
For a Bedrock target, ENABLED requires a response expression, because the model response itself never has a top-level
eventBridgefield. A feed output that runs a function with ENABLED can’t use a fixedInterval trigger shorter than 4 seconds.Possible values:
ENABLEDDISABLED
--tags (map)
The tags to associate with this function.
key -> (string)
Constraints:
- min:
1- max:
128value -> (string)
Constraints:
- min:
0- max:
256
Shorthand Syntax:
KeyName1=string,KeyName2=string
JSON Syntax:
{"string": "string"
...}
--cli-input-json | --cli-input-yaml (string)
Reads arguments from the JSON string provided. The JSON string follows the format provided by --generate-cli-skeleton. If other arguments are provided on the command line, those values will override the JSON-provided values. It is not possible to pass arbitrary binary values using a JSON-provided value as the string will be taken literally. This may not be specified along with --cli-input-yaml.
--generate-cli-skeleton (string)
Prints a JSON skeleton to standard output without sending an API request. If provided with no value or the value input, prints a sample input JSON that can be used as an argument for --cli-input-json. Similarly, if provided yaml-input it will print a sample input YAML that can be used with --cli-input-yaml. If provided with the value output, it validates the command inputs and returns a sample output JSON for that command. The generated JSON skeleton is not stable between versions of the AWS CLI and there are no backwards compatibility guarantees in the JSON skeleton generated.
--debug (boolean)
Turn on debug logging.
--endpoint-url (string)
Override command’s default URL with the given URL.
--no-verify-ssl (boolean)
By default, the AWS CLI uses SSL when communicating with AWS services. For each SSL connection, the AWS CLI will verify SSL certificates. This option overrides the default behavior of verifying SSL certificates.
--no-paginate (boolean)
Disable automatic pagination. If automatic pagination is disabled, the AWS CLI will only make one call, for the first page of results.
--output (string)
The formatting style for command output.
--query (string)
A JMESPath query to use in filtering the response data.
--profile (string)
Use a specific profile from your credential file.
--region (string)
The region to use. Overrides config/env settings.
--version (string)
Display the version of this tool.
--color (string)
Turn on/off color output.
--no-sign-request (boolean)
Do not sign requests. Credentials will not be loaded if this argument is provided.
--ca-bundle (string)
The CA certificate bundle to use when verifying SSL certificates. Overrides config/env settings.
--cli-read-timeout (int)
The maximum socket read time in seconds. If the value is set to 0, the socket read will be blocking and not timeout. The default value is 60 seconds.
--cli-connect-timeout (int)
The maximum socket connect time in seconds. If the value is set to 0, the socket connect will be blocking and not timeout. The default value is 60 seconds.
--cli-binary-format (string)
The formatting style to be used for binary blobs. The default format is base64. The base64 format expects binary blobs to be provided as a base64 encoded string. The raw-in-base64-out format preserves compatibility with AWS CLI V1 behavior and binary values must be passed literally. When providing contents from a file that map to a binary blob fileb:// will always be treated as binary and use the file contents directly regardless of the cli-binary-format setting. When using file:// the file contents will need to properly formatted for the configured cli-binary-format.
--no-cli-pager (boolean)
Disable cli pager for output.
--cli-auto-prompt (boolean)
Automatically prompt for CLI input parameters.
--no-cli-auto-prompt (boolean)
Disable automatically prompt for CLI input parameters.
--cli-error-format (string)
The formatting style for error output. By default, errors are displayed in enhanced format.
--help (boolean)
Display help for the command/subcommand.
name -> (string)
The name of the function.
Constraints:
- pattern:
[a-zA-Z0-9]([a-zA-Z0-9-_]{0,126}[a-zA-Z0-9])?
target -> (tagged union structure)
The destination for the request that the function builds.
Note
This is a Tagged Union structure. Only one of the following top level keys can be set:bedrock,passthrough.bedrock -> (structure)
Send the request to an Amazon Bedrock model.
api -> (string) [required]
The Amazon Bedrock API to call. The only valid value is INVOKE_MODEL.
Possible values:
INVOKE_MODELmodel -> (string) [required]
The model to invoke. Specify one of the following:
- A model ID.
- An inference profile ID.
- The ARN of a custom model.
- The ARN of a provisioned model.
Elemental Inference invokes the model in the Region of the feed. To use a model in other Regions, specify a cross-Region inference profile ID. A custom or provisioned model must be in the Region of the feed.
The request expression must produce a request body that this model accepts, and the model must accept the kind of media that the request expression references. For example, a text-only model can’t accept a request that references media.frames.
Constraints:
- min:
1- max:
2048- pattern:
([a-z0-9-]{1,63}[.][a-z0-9-.:]{1,76}|arn:aws:bedrock:[a-z0-9-]{1,20}:[0-9]{12}:(custom-model|provisioned-model)/.+)passthrough -> (structure)
Send the request nowhere. Use this target to reshape the metadata that Elemental Inference already produces, with no model invocation and no model charges.
requestExpression -> (string)
The JSONata expression that builds the request. Elemental Inference evaluates it against the media object of the feed, which holds the media and metadata that the function asked for.
Constraints:
- min:
1- max:
5120
responseExpression -> (string)
The JSONata expression that selects what to store. Elemental Inference evaluates it against the result as its root: the model response for a Bedrock target, or the output of the request expression for a passthrough target. If a function has no response expression, Elemental Inference stores the entire result.
Constraints:
- min:
0- max:
5120
description -> (string)
The description of the function.
Constraints:
- min:
0- max:
1024- pattern:
[\w \-\.',@:;]*
eventBridgeDelivery -> (string)
Indicates whether Elemental Inference also sends this function’s results to Amazon EventBridge. With ENABLED, each result that is a JSON object with a top-level
eventBridgefield set totrueis also sent as a Function Metadata Generated event to the default event bus in your account.Possible values:
ENABLEDDISABLED
arn -> (string)
The ARN of the function.
id -> (string)
The ID of the function.
Constraints:
- pattern:
[a-z0-9]{19}
status -> (string)
The status of the function. A successful request returns CREATING. The function is ready to use as soon as the request succeeds.
Possible values:
CREATINGAVAILABLEREFERENCEDDELETINGDELETED
references -> (list)
The IDs of the feeds that have a function output, enabled or disabled, that references this function. While this list isn’t empty, the function can’t be deleted, and its name and target can’t be changed.
(string)
Constraints:
- pattern:
[a-z0-9]{19}
contextTypes -> (list)
The kinds of media context that this function uses. Elemental Inference derives this field from the media paths that the request expression references; you can’t set it. An empty list means that the expression references no media, so Elemental Inference supplies none at run time.
Constraints:
- min:
0- max:
10(string)
One kind of media context that a function reads. Elemental Inference derives this value from the media paths that the request expression references:
- FRAMES – Derived from media.frames.
- SUBTITLES – Derived from media.transcript.
- CONTEXTUAL_METADATA – Derived from media.contextual.
- CLIP_METADATA – Derived from media.clip.
A request expression can reference at most one of SUBTITLES, CONTEXTUAL_METADATA, and CLIP_METADATA, in addition to FRAMES.
Possible values:
FRAMESSUBTITLESCONTEXTUAL_METADATACLIP_METADATA
tags -> (map)
The tags associated with the function.
key -> (string)
Constraints:
- min:
1- max:
128value -> (string)
Constraints:
- min:
0- max:
256