prompted-evaluator

PromptedEvaluator Operator Reference

PromptedEvaluator uses an LLM to score each row of text (1-5) and writes the score into a new column without deleting any rows.

1. Import

from dataflow.operators.core_text import PromptedEvaluator

2. Constructor

PromptedEvaluator(
    llm_serving=llm_serving,
    system_prompt="Please evaluate the quality of this text on a scale from 1 to 5.",
)

3. run() Signature

op.run(
    storage=self.storage.step(),
    input_key="raw_content",
    output_key="eval",
)
# returns: output_key string

4. Usage Example

from dataflow.operators.core_text import PromptedEvaluator
from dataflow.serving import APILLMServing_request
from dataflow.utils.storage import FileStorage

class MyPipeline:
    def __init__(self):
        self.storage = FileStorage(
            first_entry_file_name="./data/input.jsonl",
            cache_path="./cache",
            file_name_prefix="step",
            cache_type="jsonl"
        )

        self.llm_serving = APILLMServing_request(
            api_url="https://api.openai.com/v1/chat/completions",
            key_name_of_api_key="DF_API_KEY",
            model_name="gpt-4o",
            max_workers=10
        )

        self.evaluator = PromptedEvaluator(
            llm_serving=self.llm_serving,
            system_prompt="Evaluate text quality on a scale from 1 to 5."
        )

    def forward(self):
        self.evaluator.run(
            storage=self.storage.step(),
            input_key="content",
            output_key="quality_score"
        )

if __name__ == "__main__":
    pipeline = MyPipeline()
    pipeline.forward()