The ReasoningAnswerFormatterFilter is an operator designed to validate and filter answers based on their format. It primarily checks if a generated answer conforms to a specific structure, such as including the final answer within a \boxed{} notation, which is common in mathematical problems.

__init__ function

def __init__(self)

This operator does not require any parameters during initialization.

Prompt Template Descriptions

Prompt Template Name	Primary Purpose	Applicable Scenarios	Feature Description

run function

def run(self, storage: DataFlowStorage, input_key: str = "generated_cot")

Executes the main logic of the operator. It reads a dataframe from storage, filters rows based on the answer format in the specified input_key column, and writes the filtered dataframe back to storage.

Parameters

Name	Type	Default Value	Description
storage	DataFlowStorage	Required	The DataFlow storage instance for reading the input dataframe and writing the filtered result.
input_key	str	"generated_cot"	The name of the column containing the generated answer to be validated.

🧠 Example Usage

from dataflow.operators.reasoning import ReasoningAnswerFormatterFilter
from dataflow.utils.storage import FileStorage
from dataflow.core import LLMServingABC

class ReasoningAnswerFormatterFilterTest():
    def __init__(self, llm_serving: LLMServingABC = None):
        
        self.storage = FileStorage(
            first_entry_file_name="example.json",
            cache_path="./cache_local",
            file_name_prefix="dataflow_cache_step",
            cache_type="jsonl",
        )
        
        self.operator = ReasoningAnswerFormatterFilter()
        
    def forward(self):
        self.operator.run(
            storage = self.storage.step(),
            input_key = "output",
        )

if __name__ == "__main__":
    pl = ReasoningAnswerFormatterFilterTest()
    pl.forward()

🧾 Output Format

The operator filters the input dataframe, retaining only the rows where the answer in the input_key column passes the format validation. The schema of the output dataframe is identical to the input dataframe, but it may contain fewer rows. The filtered data is written to a new file in the storage.