17231de5d0
<!-- .github/pull_request_template.md --> ## Description <!-- Provide a clear description of the changes in this PR --> ## DCO Affirmation I affirm that all code in every commit of this pull request conforms to the terms of the Topoteretes Developer Certificate of Origin <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit - **Tests** - Updated evaluation checks by removing assertions related to the relationship between `corpus_list` and `qa_pairs`, now focusing solely on `qa_pairs` limits. - **Refactor** - Improved content processing to append each paragraph individually to `corpus_list`, enhancing clarity in data structure. - Simplified type annotations in the `load_corpus` method across multiple adapters, ensuring consistency in return types. - **Chores** - Updated dependency installation commands in GitHub Actions workflows for Python 3.10, 3.11, and 3.12 to include additional evaluation-related dependencies. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Vasilije <8619304+Vasilije1990@users.noreply.github.com>
49 lines
1.6 KiB
Python
49 lines
1.6 KiB
Python
import requests
|
|
import os
|
|
import json
|
|
import random
|
|
from typing import Optional, Any
|
|
from evals.eval_framework.benchmark_adapters.base_benchmark_adapter import BaseBenchmarkAdapter
|
|
|
|
|
|
class TwoWikiMultihopAdapter(BaseBenchmarkAdapter):
|
|
dataset_info = {
|
|
"filename": "2wikimultihop_dev.json",
|
|
"URL": "https://huggingface.co/datasets/voidful/2WikiMultihopQA/resolve/main/dev.json",
|
|
}
|
|
|
|
def load_corpus(
|
|
self, limit: Optional[int] = None, seed: int = 42
|
|
) -> tuple[list[str], list[dict[str, Any]]]:
|
|
filename = self.dataset_info["filename"]
|
|
|
|
if os.path.exists(filename):
|
|
with open(filename, "r", encoding="utf-8") as f:
|
|
corpus_json = json.load(f)
|
|
else:
|
|
response = requests.get(self.dataset_info["URL"])
|
|
response.raise_for_status()
|
|
corpus_json = response.json()
|
|
|
|
with open(filename, "w", encoding="utf-8") as f:
|
|
json.dump(corpus_json, f, ensure_ascii=False, indent=4)
|
|
|
|
if limit is not None and 0 < limit < len(corpus_json):
|
|
random.seed(seed)
|
|
corpus_json = random.sample(corpus_json, limit)
|
|
|
|
corpus_list = []
|
|
question_answer_pairs = []
|
|
for dict in corpus_json:
|
|
for title, sentences in dict["context"]:
|
|
corpus_list.append(" ".join(sentences))
|
|
|
|
question_answer_pairs.append(
|
|
{
|
|
"question": dict["question"],
|
|
"answer": dict["answer"].lower(),
|
|
"type": dict["type"],
|
|
}
|
|
)
|
|
|
|
return corpus_list, question_answer_pairs
|