Instructions to use lerobot/g05_base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use lerobot/g05_base 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
maximellerbach HF Staff
Switch to plain z-score normalization and drop the tail step
cc6410c verified Download policy_postprocessor.json from lerobot/g05_base: direct link, hf CLI and curl.
- Browser
- Download file 1.03 kB
-
https://huggingface.co/lerobot/g05_base/resolve/main/policy_postprocessor.json
- Command line
-
hf download hf://lerobot/g05_base/policy_postprocessor.json
-
curl -L -o policy_postprocessor.json https://huggingface.co/lerobot/g05_base/resolve/main/policy_postprocessor.json
1.03 kB
| { | |
| "name": "policy_postprocessor", | |
| "steps": [ | |
| { | |
| "registry_name": "g05_stepwise_unnormalizer", | |
| "config": { | |
| "eps": 1e-08, | |
| "features": { | |
| "action": { | |
| "type": "ACTION", | |
| "shape": [ | |
| 27 | |
| ] | |
| } | |
| }, | |
| "norm_map": { | |
| "ACTION": "MEAN_STD" | |
| }, | |
| "n_action_steps": 16 | |
| }, | |
| "state_file": "policy_postprocessor_step_0_g05_stepwise_unnormalizer.safetensors" | |
| }, | |
| { | |
| "registry_name": "g05_inverse_action_projection", | |
| "config": { | |
| "embodiment": "galaxea_r1lite", | |
| "policy_action_dim": 27 | |
| } | |
| }, | |
| { | |
| "registry_name": "device_processor", | |
| "config": { | |
| "device": "cpu", | |
| "float_dtype": null | |
| } | |
| }, | |
| { | |
| "registry_name": "absolute_actions_processor", | |
| "config": { | |
| "enabled": true | |
| } | |
| }, | |
| { | |
| "registry_name": "g05_action_history_crop", | |
| "config": { | |
| "num_obs_steps": 6 | |
| } | |
| } | |
| ] | |
| } |