File size: 644 Bytes
a8dcbe5 1085b14 a8dcbe5 e0dacfa 1085b14 e0dacfa a8dcbe5 1085b14 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 |
from datasets import load_dataset
def main():
raw_data = load_dataset("amazon_reviews_multi", "de")
raw_data = raw_data.rename_column("review_id", "id")
raw_data = raw_data.rename_column("review_body", "text")
raw_data = raw_data.remove_columns(["product_id", "reviewer_id", "review_title", "language", "product_category"])
for split, dataset in raw_data.items():
dataset = dataset.map(lambda x: {"label": x["stars"]-1}, num_proc=4)
dataset = dataset.map(lambda x: {"label_text": str(x["label"])}, num_proc=4)
dataset = dataset.remove_columns(["stars"])
dataset.to_json(f"{split}.jsonl")
if __name__ == "__main__":
main()
|