Download scripts/upload_hf.py from LiangYan3612/guage: direct link, hf CLI and curl.
- Browser
- Download file 792 Bytes
-
https://huggingface.co/LiangYan3612/guage/resolve/main/scripts/upload_hf.py
- Command line
-
hf download hf://LiangYan3612/guage/scripts/upload_hf.py
-
curl -L -o upload_hf.py https://huggingface.co/LiangYan3612/guage/resolve/main/scripts/upload_hf.py
792 Bytes
| # python -m verl.model_merger merge \ | |
| # --backend fsdp \ | |
| # --local_dir /workspace/AgentRL/sft_ckpt/global_step_56 \ | |
| # --target_dir /workspace/Qwen-4B-webshop | |
| from huggingface_hub import create_repo, upload_folder | |
| # 设置模型名称(在你 HuggingFace 账号下) | |
| repo_name = "willamazon1/Qwen3-4B-webshop-gpt" | |
| # 可选:如果模型属于某个组织(例如 huggingface/your-org-model) | |
| # repo_id = "your-org/my-sft-model" | |
| repo_id = repo_name # 若是上传到你自己账号 | |
| # 创建仓库(自动跳过已存在) | |
| create_repo(repo_id, exist_ok=True) | |
| # 上传整个模型文件夹 | |
| upload_folder( | |
| repo_id=repo_id, | |
| folder_path="/workspace/Qwen3-4B-webshop", # 你的本地模型路径 | |
| path_in_repo=".", # 上传到仓库的根目录 | |
| ) | |