Binary Choice Agent

This Agent will analyze the image and query, and determine whether the resulting classification is of the class “yes” or “no”.

Initialization

The BinaryChoiceAgent is initialized with two arguments:

BinaryChoiceAgent(query, model(Optional))

query

String

required

Specifies the text prompt that will be used to analyze the image.

model

MultimodalLLM

Represents the model used to perform the binary classification task.

If left unspecified, this parameter defaults to the QwenVL() Model. The supported MultimodalLLM models can be found below:

example.py
BinaryChoiceAgent("Is this person wearing a hardhat?")

The output from this agent will be an ExecutionNode containing the previous image and the classification result (“yes” or “no”).

On this page