chore: publish rSkill OpenRAL/rskill-gr00t_n17-franka_panda-libero_spatial-bf16 v0.1.0
Browse files- eval/libero_spatial.json +50 -0
eval/libero_spatial.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"schema_version": "0.1",
|
| 3 |
+
"source": {
|
| 4 |
+
"paper": "https://arxiv.org/abs/2306.03310",
|
| 5 |
+
"arxiv": "https://arxiv.org/abs/2306.03310",
|
| 6 |
+
"model_variant": "gr00t",
|
| 7 |
+
"evaluated_by": "OpenRAL:openral benchmark run",
|
| 8 |
+
"reproduced_locally": true,
|
| 9 |
+
"reproduction_planned": null,
|
| 10 |
+
"reproduction_cli": "openral benchmark run --suite libero_spatial --rskill rskills/gr00t-n17-libero",
|
| 11 |
+
"table": null,
|
| 12 |
+
"status": "reproduced"
|
| 13 |
+
},
|
| 14 |
+
"benchmark": {
|
| 15 |
+
"name": "LIBERO-Spatial",
|
| 16 |
+
"dataset": null,
|
| 17 |
+
"protocol": "1 episodes per task, success_key=is_success, max_steps=280",
|
| 18 |
+
"robot": "franka_panda",
|
| 19 |
+
"simulator": "LIBERO (MuJoCo)"
|
| 20 |
+
},
|
| 21 |
+
"eval_config": {
|
| 22 |
+
"n_episodes": 1,
|
| 23 |
+
"seeds": [
|
| 24 |
+
0
|
| 25 |
+
],
|
| 26 |
+
"success_key": "is_success",
|
| 27 |
+
"max_steps": 280,
|
| 28 |
+
"vla_id": "gr00t",
|
| 29 |
+
"weights_uri": "rskills/gr00t-n17-libero"
|
| 30 |
+
},
|
| 31 |
+
"results": {
|
| 32 |
+
"libero_spatial/0_success_rate": 1.0,
|
| 33 |
+
"libero_spatial/1_success_rate": 1.0,
|
| 34 |
+
"libero_spatial/2_success_rate": 1.0,
|
| 35 |
+
"libero_spatial/3_success_rate": 1.0,
|
| 36 |
+
"libero_spatial/4_success_rate": 1.0,
|
| 37 |
+
"libero_spatial/5_success_rate": 0.0,
|
| 38 |
+
"libero_spatial/6_success_rate": 1.0,
|
| 39 |
+
"libero_spatial/7_success_rate": 1.0,
|
| 40 |
+
"libero_spatial/8_success_rate": 1.0,
|
| 41 |
+
"libero_spatial/9_success_rate": 1.0,
|
| 42 |
+
"avg_success_rate": 0.9,
|
| 43 |
+
"n_tasks": 10,
|
| 44 |
+
"n_episodes_per_task": 1,
|
| 45 |
+
"n_episodes_total": 10,
|
| 46 |
+
"mean_step_latency_ms_avg": 36.12678431379617
|
| 47 |
+
},
|
| 48 |
+
"baselines": {},
|
| 49 |
+
"trace_id": null
|
| 50 |
+
}
|