Cannot use two_hop_incremental

[boltzmann-brain]

Hello,

I was playing around with the QA dataset generation capabilities and getting an error
You have to sample more than two passages for making two-hop questions.

I am using random_single_hop, n=50 to sample from the corpus. Am I doing something wrong? I don't see double hop sampling functions; the only sampling functions available seem to be single hop.

Full code (adapted from your Evaluation data creation tutorial in the docs):

import pandas as pd
from llama_index.llms.openai import OpenAI
from openai import AsyncOpenAI

from autorag.data.qa.filter.dontknow import dontknow_filter_rule_based
from autorag.data.qa.filter.passage_dependency import passage_dependency_filter_openai
from autorag.data.qa.generation_gt.llama_index_gen_gt import (
    make_basic_gen_gt,
    make_concise_gen_gt,
)
from autorag.data.qa.schema import Raw, Corpus
from autorag.data.qa.query.llama_gen_query import factoid_query_gen, concept_completion_query_gen, two_hop_incremental
from autorag.data.qa.sample import random_single_hop

llm = OpenAI()
raw_df = pd.read_parquet("./all_txt_parsing_output/parsed_result.parquet")
raw_instance = Raw(raw_df)

corpus_df = pd.read_parquet("all_txt_chunking_output/0.parquet")
corpus_instance = Corpus(corpus_df, raw_instance)

initial_qa = (
    corpus_instance.sample(random_single_hop, n=50)
    .map(
        lambda df: df.reset_index(drop=True),
    )
    .make_retrieval_gt_contents()
    .batch_apply(
        two_hop_incremental,
        llm=llm, # query generation
    )
    .batch_apply(
        make_basic_gen_gt,  # answer generation (basic)
        llm=llm,
    )
    .batch_apply(
        make_concise_gen_gt,  # answer generation (concise)
        llm=llm,
    )
    .filter(
        dontknow_filter_rule_based,  # filter don't know
        lang="en",
    )
)

initial_qa.to_parquet('./qa_output/factoid_twohop_basic_concise_dontknow/all_docs_qa.parquet', './all_docs_corpus.parquet')