Instructions to use yrpark/elice_gr00t_install with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use yrpark/elice_gr00t_install with LeRobot:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Download upload_task.py from yrpark/elice_gr00t_install: direct link, hf CLI and curl.
- Browser
- Download file 2.51 kB
-
https://huggingface.co/yrpark/elice_gr00t_install/resolve/main/upload_task.py
- Command line
-
hf download hf://yrpark/elice_gr00t_install/upload_task.py
-
curl -L -o upload_task.py https://huggingface.co/yrpark/elice_gr00t_install/resolve/main/upload_task.py
2.51 kB
| #!/usr/bin/env python | |
| """upload_task.py — upload one trained task's pretrained_model folder to a PUBLIC | |
| model repo as a subfolder, then verify what landed. | |
| Usage: | |
| python upload_task.py <PRETRAINED_MODEL_DIR> <PATH_IN_REPO> <REPO_ID> <EXPECT_ROOT_SUBSTR> | |
| Uploads the WHOLE folder (handles sharded safetensors). Creates repo (private=False) | |
| if missing. | |
| """ | |
| import json | |
| import os | |
| import sys | |
| from huggingface_hub import HfApi, create_repo, upload_folder | |
| def main(): | |
| src, path_in_repo, repo_id, expect = sys.argv[1:5] | |
| if not os.path.isdir(src): | |
| print(f"[upload] SRC MISSING: {src} -> SKIP") | |
| sys.exit(1) | |
| create_repo(repo_id, repo_type="model", private=False, exist_ok=True) | |
| print(f"[upload] {src} -> {repo_id}:{path_in_repo} (public)") | |
| upload_folder( | |
| folder_path=src, | |
| path_in_repo=path_in_repo, | |
| repo_id=repo_id, | |
| repo_type="model", | |
| commit_message=f"Add {path_in_repo} GR00T N1.7 EE checkpoint", | |
| ) | |
| print(f"[verify] files under {repo_id}:{path_in_repo}") | |
| api = HfApi() | |
| files = [f for f in api.list_repo_files(repo_id, repo_type="model") if f.startswith(path_in_repo + "/")] | |
| total = 0 | |
| infos = api.model_info(repo_id, files_metadata=True).siblings | |
| size_by = {s.rfilename: (s.size or 0) for s in infos} | |
| for f in sorted(files): | |
| sz = size_by.get(f, 0) | |
| total += sz | |
| print(f" {f} ({sz/1e6:.1f} MB)") | |
| print(f"[verify] safetensors total ~ {total/1e9:.2f} GB") | |
| # local content checks | |
| tc = os.path.join(src, "train_config.json") | |
| cfg = os.path.join(src, "config.json") | |
| if os.path.exists(tc): | |
| t = json.load(open(tc)) | |
| d = t.get("dataset", {}) | |
| root = str(d.get("root")); rid = str(d.get("repo_id")) | |
| ok = expect in root or expect in rid | |
| print(f"[verify] train_config dataset.root={root} repo_id={rid} -> " | |
| f"{'OK points to ' + expect if ok else 'WARN does not mention ' + expect}") | |
| else: | |
| print("[verify] train_config.json MISSING locally") | |
| if os.path.exists(cfg): | |
| c = json.load(open(cfg)) | |
| print(f"[verify] config.json: model_version={c.get('model_version')} " | |
| f"base_model_path={c.get('base_model_path')} embodiment_tag={c.get('embodiment_tag')}") | |
| has_index = any(f.endswith("model.safetensors.index.json") for f in files) | |
| print(f"[verify] sharded index present: {has_index}") | |
| print(f"[verify] DONE {path_in_repo}") | |
| if __name__ == "__main__": | |
| main() | |