Comment on page

Configuration

This section explains each of the properties in the configuration blade. 
General
Property
Description
Collection
This associates the Agent to a specific Collection (default to that of the current Data Stream).
Refer to Collections and Stream Hosts to understand more about Collections.
API Server
Property
Description
Version Poll Interval (seconds)
The interval between requests to the API Server to check the published version of the model being run. Defaults to 3600 (one hour).
Use Variables
Tick to use variables for the API URL and Runtime URL fields; or untick to enter them manually.
API Url
The Url of an MLflow server that provides information about experiments and running models.
Model
The name of the model to send data to. The list is populated with the published models retrieved from the API Url.
Runtime Server
Property
Description
Runtime Url
The Url at which the MLflow model has been published.
Before configuring the AI & ML Agent, please ensure that its input endpoint is connected to a parent Agent which will be sending data to it.
Input Mapping
A parent Agent attribute is mapped against each feature in the MLflow model's signature to provide it with input data.
Ensure that a signature is logged with the model before mapping the inputs.
Refer to Setup Input Mappings for step-by-step instructions.
Endpoints
Name
Description
Input
This endpoint is used to receive data from the parent Agent.
Output
The output of the processed events are added to the input event data and made available at this endpoint.
Events received from the parent Agent are made available to this endpoint, with MLflow model's version and prediction appended to the original event.

Property	Description
Collection	This associates the Agent to a specific Collection (default to that of the current Data Stream).

Property	Description
Version Poll Interval (seconds)	The interval between requests to the API Server to check the published version of the model being run. Defaults to 3600 (one hour).
Use Variables	Tick to use variables for the API URL and Runtime URL fields; or untick to enter them manually.
API Url	The Url of an MLflow server that provides information about experiments and running models.
Model	The name of the model to send data to. The list is populated with the published models retrieved from the API Url.

Property	Description
Runtime Url	The Url at which the MLflow model has been published.

Name	Description
Input	This endpoint is used to receive data from the parent Agent.
Output	The output of the processed events are added to the input event data and made available at this endpoint. Events received from the parent Agent are made available to this endpoint, with MLflow model's version and prediction appended to the original event.

How to use? - Previous

Example

Next - How to use?

Release Notes

Last modified 4mo ago