"""Upload the tokenised data folder to the private HF dataset QLNI/embeddings-are-overpriced-data. python upload_data.py --data data """ import argparse from huggingface_hub import HfApi ap = argparse.ArgumentParser() ap.add_argument("--data", default="data") a = ap.parse_args() api = HfApi() api.create_repo("QLNI/embeddings-are-overpriced-data", repo_type="dataset", private=True, exist_ok=True) api.upload_large_folder(repo_id="QLNI/embeddings-are-overpriced-data", repo_type="dataset", folder_path=a.data, allow_patterns=["*.npy"]) print("uploaded", a.data, flush=True)