BHARGAV REDDY commited on
Upload Base/Datasets/rag_mcp_sft/push_to_hf.py with huggingface_hub
Browse files
Base/Datasets/rag_mcp_sft/push_to_hf.py
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import argparse
|
| 2 |
+
import os
|
| 3 |
+
from pathlib import Path
|
| 4 |
+
|
| 5 |
+
from huggingface_hub import HfApi
|
| 6 |
+
|
| 7 |
+
|
| 8 |
+
def main():
|
| 9 |
+
parser = argparse.ArgumentParser(description="Upload the RAG+MCP SFT dataset folder to Hugging Face")
|
| 10 |
+
parser.add_argument("--repo-id", default="ASTERIZER/LUNA-RAG-MCP-SFT-10M")
|
| 11 |
+
parser.add_argument("--folder", default=str(Path(__file__).resolve().parent))
|
| 12 |
+
parser.add_argument("--private", action="store_true")
|
| 13 |
+
args = parser.parse_args()
|
| 14 |
+
|
| 15 |
+
token = os.environ.get("HF_TOKEN")
|
| 16 |
+
if not token:
|
| 17 |
+
raise RuntimeError("HF_TOKEN is required in the environment to push the dataset")
|
| 18 |
+
|
| 19 |
+
api = HfApi(token=token)
|
| 20 |
+
api.create_repo(repo_id=args.repo_id, repo_type="dataset", private=args.private, exist_ok=True)
|
| 21 |
+
api.upload_folder(
|
| 22 |
+
repo_id=args.repo_id,
|
| 23 |
+
repo_type="dataset",
|
| 24 |
+
folder_path=args.folder,
|
| 25 |
+
path_in_repo=".",
|
| 26 |
+
)
|
| 27 |
+
print(f"uploaded_dataset repo_id={args.repo_id} folder={args.folder}")
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
if __name__ == "__main__":
|
| 31 |
+
main()
|