realtime-rag-pipeline / data /load_dataset.py
Gourisankar Padihary
Compute RMSE and AUCROC
f7c2fa3
raw
history blame
311 Bytes
import logging
from datasets import load_dataset
def load_data():
logging.info("Loading dataset")
dataset = load_dataset("rungalileo/ragbench", 'covidqa', split="test")
logging.info("Dataset loaded successfully")
logging.info(f"Number of documents found: {dataset.num_rows}")
return dataset