from datasets import load_dataset


def main():

	raw_data = load_dataset("catalonia_independence", "catalan")
	raw_data = raw_data.rename_column("LABEL", "label")
	raw_data = raw_data.rename_column("TWEET", "text")
	raw_data = raw_data.rename_column("id_str", "id")

	labels = raw_data["train"].features["label"]


	for split, dataset in raw_data.items():
		dataset = dataset.map(lambda x: {"label_text": labels.int2str(x["label"])}, num_proc=4)
		dataset.to_json(f"{split}.jsonl")

if __name__ == "__main__":
	main()