Files
DataPipeline/english-fineweb.py
T

9 lines
264 B
Python

from datasets import load_dataset
from utils import pre_tarin_process
if __name__ == "__main__":
dataset = load_dataset("HuggingFaceFW/fineweb", "sample-10BT")
pre_tarin_process(
dataset_dict=dataset,
output_subdir="english-fineweb",
)