simrig-orbit-reach / evaluation.json
ssuleiman's picture
Add orbit-arm reach PPO reference checkpoint (v1)
c469f0d verified
Raw History Blame Contribute Delete
3.53 kB
{
"suite": "promotion",
"passed": true,
"scenario": "displayed",
"episodes": 60,
"successes": 54,
"success_rate": 0.9,
"safety_failures": 0,
"failed_seeds": [
9,
10,
13,
16,
18,
43
],
"failures": [
{
"seed": 9,
"scenario": "displayed",
"terminal_reason": {
"category": "task_failure",
"code": "target_not_sustained",
"details": {
"actual": 0,
"expected": 5,
"predicate_id": "sustained_target_arrival"
},
"message": "Predicate 'sustained_target_arrival' failed: actual=0 expected=5."
},
"maximum_hold_steps": 0,
"minimum_target_error": 0.24030554329234238,
"forbidden_contact_count": 0
},
{
"seed": 10,
"scenario": "displayed",
"terminal_reason": {
"category": "task_failure",
"code": "target_not_sustained",
"details": {
"actual": 0,
"expected": 5,
"predicate_id": "sustained_target_arrival"
},
"message": "Predicate 'sustained_target_arrival' failed: actual=0 expected=5."
},
"maximum_hold_steps": 0,
"minimum_target_error": 0.22367826451961925,
"forbidden_contact_count": 0
},
{
"seed": 13,
"scenario": "displayed",
"terminal_reason": {
"category": "task_failure",
"code": "target_not_sustained",
"details": {
"actual": 0,
"expected": 5,
"predicate_id": "sustained_target_arrival"
},
"message": "Predicate 'sustained_target_arrival' failed: actual=0 expected=5."
},
"maximum_hold_steps": 0,
"minimum_target_error": 0.10712519091374702,
"forbidden_contact_count": 0
},
{
"seed": 16,
"scenario": "displayed",
"terminal_reason": {
"category": "task_failure",
"code": "target_not_sustained",
"details": {
"actual": 0,
"expected": 5,
"predicate_id": "sustained_target_arrival"
},
"message": "Predicate 'sustained_target_arrival' failed: actual=0 expected=5."
},
"maximum_hold_steps": 0,
"minimum_target_error": 0.0926555449924868,
"forbidden_contact_count": 0
},
{
"seed": 18,
"scenario": "displayed",
"terminal_reason": {
"category": "task_failure",
"code": "target_not_sustained",
"details": {
"actual": 0,
"expected": 5,
"predicate_id": "sustained_target_arrival"
},
"message": "Predicate 'sustained_target_arrival' failed: actual=0 expected=5."
},
"maximum_hold_steps": 0,
"minimum_target_error": 0.1542196887033212,
"forbidden_contact_count": 0
},
{
"seed": 43,
"scenario": "displayed",
"terminal_reason": {
"category": "task_failure",
"code": "target_not_sustained",
"details": {
"actual": 0,
"expected": 5,
"predicate_id": "sustained_target_arrival"
},
"message": "Predicate 'sustained_target_arrival' failed: actual=0 expected=5."
},
"maximum_hold_steps": 0,
"minimum_target_error": 0.10512675214180775,
"forbidden_contact_count": 0
}
],
"trainer_eval_episode_success": 0.8125,
"checkpoint_sha256": "8874731b50eca8b649df6b63b151526a8f2ba2dfd6276719e69858be9d2d26eb",
"config_sha256": "f4c5ded83e7b696ef84e365067d560509b18095cca0ec9ed4357ce0309c56ae2"
}