Add FPO checkpoint metadata
Browse files
models/fpo_mustard_dexpolicy_seed0.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"method": "FPO + DexPolicy",
|
| 3 |
+
"object": "ycb-006_mustard_bottle",
|
| 4 |
+
"seed": 0,
|
| 5 |
+
"training_budget_steps": 5000000,
|
| 6 |
+
"learning_rate": 0.00001,
|
| 7 |
+
"clip_range": 0.20,
|
| 8 |
+
"update_epochs": 5,
|
| 9 |
+
"flow_integration_steps": 8,
|
| 10 |
+
"exploration_schedule": "linear 0.20 to 0.05 over 5M steps"
|
| 11 |
+
}
|