Upload upload_full_sft_to_hf.py with huggingface_hub

upload_full_sft_to_hf.py

@@ -0,0 +1,57 @@
+import argparse
+import os
+from pathlib import Path
+
+from huggingface_hub import HfApi
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(description="Upload a full SFT run folder to a Hugging Face model repo")
+    parser.add_argument("--repo-id", default="ASTERIZER/LUNA-100M")
+    parser.add_argument("--folder", default="Base/out/sft/rag_mcp_full_sft")
+    parser.add_argument("--path-in-repo", default="rag_mcp_full_sft")
+    parser.add_argument("--private", action="store_true")
+    return parser.parse_args()
+
+
+def main():
+    args = parse_args()
+    token = os.environ.get("HF_TOKEN")
+    if not token:
+        raise RuntimeError("HF_TOKEN is required in the environment to upload the full SFT folder")
+
+    folder = Path(args.folder)
+    if not folder.exists():
+        raise FileNotFoundError(f"Full SFT output folder not found: {folder}")
+    if not folder.is_dir():
+        raise NotADirectoryError(f"Expected a folder, got: {folder}")
+
+    required_files = [folder / "final" / "model.pth", folder / "latest.pt"]
+    missing = [str(path) for path in required_files if not path.exists()]
+    if missing:
+        raise FileNotFoundError("Missing expected full SFT files: " + ", ".join(missing))
+
+    api = HfApi(token=token)
+    api.create_repo(
+        repo_id=args.repo_id,
+        repo_type="model",
+        private=args.private,
+        exist_ok=True,
+    )
+    api.upload_folder(
+        repo_id=args.repo_id,
+        repo_type="model",
+        folder_path=str(folder),
+        path_in_repo=args.path_in_repo,
+    )
+    print(
+        "uploaded_full_sft "
+        f"repo_id={args.repo_id} "
+        f"folder={folder} "
+        f"path_in_repo={args.path_in_repo} "
+        f"url=https://huggingface.co/{args.repo_id}/tree/main/{args.path_in_repo}"
+    )
+
+
+if __name__ == "__main__":
+    main()