Unverified Commit f8c2cfcb authored by Hanwool Albert Lee's avatar Hanwool Albert Lee Committed by GitHub
Browse files

Merge pull request #1089 from h-albert-lee/K-MMLU

Add kmmlu evaluation to tasks
parents c86f527f c256eda8
"dataset_name": "Public Safety"
"include": "_default_kmmlu_yaml"
"task": "kmmlu_public_safety"
"dataset_name": "Railway and Automotive Engineering"
"include": "_default_kmmlu_yaml"
"task": "kmmlu_railway_and_automotive_engineering"
"dataset_name": "Real Estate"
"include": "_default_kmmlu_yaml"
"task": "kmmlu_real_estate"
"dataset_name": "Refrigerating Machinery"
"include": "_default_kmmlu_yaml"
"task": "kmmlu_refrigerating_machinery"
"dataset_name": "Social Welfare"
"include": "_default_kmmlu_yaml"
"task": "kmmlu_social_welfare"
"dataset_name": "Taxation"
"include": "_default_kmmlu_yaml"
"task": "kmmlu_taxation"
"dataset_name": "Telecommunications and Wireless Technology"
"include": "_default_kmmlu_yaml"
"task": "kmmlu_telecommunications_and_wireless_technology"
import datasets
def process_docs(dataset: datasets.Dataset) -> datasets.Dataset:
def _process_doc(doc):
instruction = (
f"다음을 읽고 정답으로 알맞은 것을 고르시요.\n"
f"### Question: {doc['question']}\n"
f"### Options:\n"
f"(1) {doc['option#1']}\n(2) {doc['option#2']}\n(3) {doc['option#3']}\n(4) {doc['option#4']}\n"
f"### Answer: 주어진 문제의 정답은"
)
out_doc = {
"question": instruction,
"choices": ["(1)", "(2)", "(3)", "(4)"],
"gold": int(doc["answer"]) - 1,
}
return out_doc
return dataset.map(_process_doc)
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment