Views
No views yet
dynamics_model/ # Stage 1: Learned dynamics models
├── beamng_bicycle_*/ # Bicycle model variants
├── beamng_ddm_*/ # Deep Dynamics Model (DDM) variants
├── beamng_trans_*/ # Transformer-based models
├── beamng_dytr_*/ # DYTR (Residual Learning) models
└── beamng_manual_PID_*/ # Manual PID control baselines
control_policies/ # Stage 2 & 3: Trained RL policies
├── PPO____*_bicycle/ # Policies trained on bicycle dynamics
├── PPO____*_ddm/ # Policies trained on DDM dynamics
├── PPO____*_trans/ # Policies trained on Transformer dynamics
├── PPO____*_dytr_ddm/ # Policies trained on DYTR dynamics
└── PPO____*_oracle/ # Oracle policies (full-state observation)1from huggingface_hub import hf_hub_download
2
3# Download a dynamics model
4dynamics_model = hf_hub_download(
5 repo_id="alfredgu001324/Sim2Sim2Sim",
6 filename="dynamics_model/beamng_trans_10/best_model.pt"
7)
8
9# Download a trained policy
10policy = hf_hub_download(
11 repo_id="alfredgu001324/Sim2Sim2Sim",
12 filename="control_policies/PPO____R_80000__11_25_10_41_44_840_trans/model_PPO____R_80000__11_25_10_41_44_840_001280.pt"
13)1from huggingface_hub import snapshot_download
2
3# Download all checkpoints
4local_dir = snapshot_download(
5 repo_id="alfredgu001324/Sim2Sim2Sim",
6 cache_dir="./ckpts",
7 repo_type="model"
8)1@article{GuChittaEtAl2026,
2 author = {Gu, Xunjiang and Chitta, Kashyap and Golchoubian, Mahsa and Suplin, Vladimir and Gilitschenski, Igor},
3 title = {Dynamics Distillation for Efficient and Transferable Control Learning},
4 journal = {arXiv preprint arXiv:2605.01516},
5 year = {2026}
6}