iFaz/diffusion-libero_goal-seed3-15000

DIFFUSION Policy — diffusion_libero_goal_seed3

Trained with LeRobot.
Date: 2026-05-18 06:13
Policy type: diffusion | Device: cuda

---

📦 Dataset

Parameter	Value
dataset.repo_id	HuggingFaceVLA/libero

---

🏋️ Training Config

Parameter	Value
steps	15000
batch_size	8
eval_freq	0
save_freq	3000
num_workers	4
seed	3
eval.n_episodes	1
eval.batch_size	1
eval.use_async_envs	True

---

📐 Policy Architecture

All defaults — no overrides applied.

---

🎯 Eval Config

Parameter	Value
env.type	libero
env.task	libero_goal
eval.n_episodes	1
eval.batch_size	4
eval.use_async_envs	False
policy.path	/kaggle/working/outputs/train/libero_goal_seed3/checkpoints/last/pretrained_model

---

📊 Eval Results

Metric	Value
Episodes	10
Success rate	0.0%
Avg sum reward	0.00
Avg max reward	0.00
Eval time (s)	1355.6

---

Citation

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