iFaz/eqm-aloha_transfer_cube-seed3-18sep2026_3pm

EQM Policy — eqm_aloha_transfer_cube_seed3_18sep2026_3pm

Trained with LeRobot. Date: 2026-09-18 16:04 Policy type: eqm | Device: cuda

---

📦 Dataset

Parameter	Value
dataset.repo_id	lerobot/aloha_sim_transfer_cube_human

---

🏋️ Training Config

Parameter	Value
steps	2000
batch_size	8
eval_freq	0
save_freq	1000
num_workers	4
seed	3
eval.n_episodes	1
eval.batch_size	1
eval.use_async_envs	True

---

📐 Policy Architecture

Parameter	Value
ebm	dot
jacobian_reg_weight	0.0001
jacobian_reg_probes	4
fixed_point_anchor_weight	0.1
c_gamma_a	0.5
use_adaptive_compute	False

---

🎯 Eval Config

Parameter	Value
env.type	aloha
env.task	AlohaTransferCube-v0
eval.n_episodes	5
eval.batch_size	1
eval.use_async_envs	False
policy.path	/content/outputs/train/aloha_transfer_cube_seed3/checkpoints/last/pretrained_model
policy.ood_logging_enabled	True
policy.ood_calibration_stats_path	/content/eqm_calibration.json
policy.ood_log_path	/content/outputs/eqm_ood_log.csv
policy.ood_z_threshold	3.0
policy.sample_stepsize	0.17

---

📊 Eval Results

Metric	Value
Episodes	5
Success rate	20.0%
Avg sum reward	0.80
Avg max reward	0.80
Eval time (s)	177.4

---

Citation

@misc{cadene2024lerobot,
  author = {Cadene, Remi and Alibert, Simon and others},
  title  = {LeRobot},
  year   = {2024},
  url    = {https://github.com/huggingface/lerobot}
}