diff --git a/checkpoints/210122235754-5000.pth.actor b/checkpoints/210122235754-5000.pth.actor new file mode 100644 index 0000000000000000000000000000000000000000..07661e685ff2f34e9320d6a83f6cac3f9629807e Binary files /dev/null and b/checkpoints/210122235754-5000.pth.actor differ diff --git a/checkpoints/210122235754-5000.pth.optimizer b/checkpoints/210122235754-5000.pth.optimizer new file mode 100644 index 0000000000000000000000000000000000000000..d581eb40d2a7e6dccc55697a4676883cc01d85c2 Binary files /dev/null and b/checkpoints/210122235754-5000.pth.optimizer differ diff --git a/checkpoints/210122235754-5000.pth.value b/checkpoints/210122235754-5000.pth.value new file mode 100644 index 0000000000000000000000000000000000000000..c323fa9a74a8e6c1ba62b0140fbebaa6e85b824b Binary files /dev/null and b/checkpoints/210122235754-5000.pth.value differ diff --git a/run.py b/run.py index 6e2a3d88c4945b6b3a9c7159e7ba7f965ffc20cc..f4c75dbc19976ea7758966fc326a00010a9b38ab 100644 --- a/run.py +++ b/run.py @@ -71,6 +71,15 @@ load_policy = "DDDQN" checkpoint = "./checkpoints/210122120236-3000.pth" # 17.011131341978228 EPSILON = 0.0 + +load_policy = "DDDQN" +checkpoint = "./checkpoints/210122165109-5000.pth" # 14.438561289066921 +EPSILON = 0.0 + +load_policy = "PPO" +checkpoint = "./checkpoints/210122235754-5000.pth" # 16.2589510774425 +EPSILON = 0.0 + # load_policy = "DeadLockAvoidance" # 22.13346834815911 # Use last action cache