📘 Overview

PairedPromptedGenerator is a paired-input prompt generator. It is designed to take two columns from a dataset, combine them with a system prompt to create a complete prompt, and then use a Large Language Model (LLM) to generate a response. The operator processes data in batches and writes the generated content back to a new column.

__init__

def __init__(self, llm_serving: LLMServingABC, system_prompt: str = "You are a helpful agent.")

Parameter	Type	Default Value	Description
llm_serving	LLMServingABC	Required	The Large Language Model serving instance used for inference and generation.
system_prompt	str	"You are a helpful agent."	A system-level instruction that prefixes each sample to guide the model's role and output style.

Prompt Template Descriptions

Prompt Template Name	Primary Use	Applicable Scenarios	Feature Description

run

def run(self, storage: DataFlowStorage, input_key_1: str = "input_key_1", input_key_2: str = 'input_key_2', output_key: str = "generated_content")

Executes the main logic of the operator. It reads an input DataFrame from storage, generates LLM-based responses from paired input columns, and writes the results back to storage.

Parameters

Name	Type	Default Value	Description
storage	DataFlowStorage	Required	The data flow storage instance for reading and writing data.
input_key_1	str	"input_key_1"	The name of the column for the first part of the paired input.
input_key_2	str	"input_key_2"	The name of the column for the second part of the paired input.
output_key	str	"generated_content"	The name of the column where the generated output will be stored.

🧠 Example Usage

🧾 Default Output Format

Field	Type	Description
(original fields)	-	The original fields from the input data are preserved.
generated_content	str	The response generated by the model based on the paired inputs.

Example Input:

{
  "input_key_1": "Translate the following English text to French:",
  "input_key_2": "Data science is an interdisciplinary field."
}

Example Output:

{
  "input_key_1": "Translate the following English text to French:",
  "input_key_2": "Data science is an interdisciplinary field.",
  "generated_content": "La science des données est un domaine interdisciplinaire."
}