SkillMimicwyhuai/SkillMimic
Provides a configurable training and evaluation entry point for physics-simulation-based humanoid motion imitation learn…
No ratings yet
Rate this skill
Stars445
Forks28
Watchers445
Open Issues4
InstallOne-line install
UnconfirmedWhether this repository follows the Agent Skills spec (that is, whether it contains a SKILL.md) has not been verified yet. Until that is settled, clone it and use what you need:
git clone https://github.com/wyhuai/SkillMimic.git
OverviewOverview
Key Features
Provides a configurable training and evaluation entry point for physics-simulation-based humanoid motion imitation learning, controlling environments, training pipelines, and motion data via command-line arguments. Key capabilities include: 1. Environment config loading: specify simulation environment configuration files via --cfg_env to quickly switch robot models and scene parameters; 2. Training config loading: specify training hyperparameters and algorithm configurations via --cfg_train to reproduce or adjust training pipelines; 3. Motion data specification: specify reference motion files via --motion_file so the robot imitates target motion trajectories; 4. Headless mode execution: disable the graphical interface via --headless to run training and evaluation in batches on servers; 5. Policy checkpoint loading: load trained low-level controller weights via --llc_checkpoint for continued training or inference evaluation.
Use Cases
Training and tuning humanoid motion imitation algorithms, quick comparison of simulation environments, batch experiments on headless servers, policy fine-tuning and evaluation based on existing checkpoints.
Target Roles
AI / Algorithm Engineer, Hardware / Embedded Engineer, Test / QA Engineer.
RatingUser ratings
Nobody has rated this skill yet. Yours would be the first on the site — your vote becomes the initial average.
RepositoryRepository
| Skill ID | 1911 |
|---|---|
| Data source | GitHub |
| Repository | wyhuai/SkillMimic |
| Category | Artificial Intelligence & Machine Learning / Reinforcement Learning & Motion Control |
| Primary language | Python |
| License | Apache License 2.0 |
| Default branch | main |
| GitHub repository ID | 836765735 |
| Node ID | R_kgDOMeAIJw |
| Rank | #1416 (position within the source ranking) |
| Site-wide popularity rank | #1609 (computed live from Stars, across all 5833 entries) |
GrowthStar / Fork growth
One snapshot per day, tracking how Stars and Forks changed
- Now445
- Previous point · 09-28 443
- vs previous point +2
- 2 data points 13 days
- Now28
- Previous point · 09-28 28
- vs previous point ±0
- 2 data points 13 days
TimelineTimeline
Repository created2024-08-01 14:08:072024-08-01
Data last updated2026-10-26 12:13:5723 days ago
Last commit2025-06-30 06:59:502025-06-30
CommitLatest commit
Version [latest] committed by QihanZhao. Update Category: Documentation & Guidelines Update. Commit details: Update README.md. This update improves usage instructions, deployment guides, and developer docs.