Run v3

training/runs/v3/config.json

@@ -0,0 +1,15 @@
+{
+  "W_PER_SHEEP": 2.0,
+  "W_ALIGN": 0.05,
+  "W_PEN_BONUS": 10.0,
+  "W_COMPLETE": 100.0,
+  "W_STEP_COST": 0.02,
+  "W_SOUTH": 0.01,
+  "W_COMPACT": 0.0,
+  "W_WALL_TOUCH": 0.0,
+  "WALL_TOUCH_BUFFER": 0.4,
+  "ALIGN_SHAPE": "standoff",
+  "ALIGN_GATED": true,
+  "ENTRY_AWARE": true,
+  "ent_coef": 0.02
+}

training/runs/v3/stage_results.json

@@ -0,0 +1,222 @@
+[
+  {
+    "sr": 1.0,
+    "mean_len": 264.3,
+    "mean_min_pen": 3.6947483142217,
+    "mean_act": 0.4488927691353647,
+    "failure_modes": {
+      "SUCCESS": 30
+    },
+    "reward_per_step": {
+      "progress": 0.11562251145796992,
+      "alignment": 0.00012847888517811197,
+      "south": -0.00046327802870008703,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.037835792659856225,
+      "step_cost": -0.020000000000000923,
+      "complete": 0.37835792659856227
+    },
+    "n_sheep": 1
+  },
+  {
+    "sr": 0.7666666666666667,
+    "mean_len": 1397.6333333333334,
+    "mean_min_pen": 3.3354002753893535,
+    "mean_act": 0.9679237489606706,
+    "failure_modes": {
+      "SUCCESS": 23,
+      "PARTIAL_1of2": 6,
+      "COMPACT_CANT_DRIVE": 1
+    },
+    "reward_per_step": {
+      "progress": 0.04012407340533507,
+      "alignment": 0.004549029322963513,
+      "south": -0.003855391958439705,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.01264041594123399,
+      "step_cost": -0.019999999999988728,
+      "complete": 0.05485463521667581
+    },
+    "n_sheep": 2
+  },
+  {
+    "sr": 0.9666666666666667,
+    "mean_len": 1095.3666666666666,
+    "mean_min_pen": 2.4724439779917398,
+    "mean_act": 0.950618689999602,
+    "failure_modes": {
+      "SUCCESS": 29,
+      "COMPACT_CANT_DRIVE": 1
+    },
+    "reward_per_step": {
+      "progress": 0.08207998032411863,
+      "alignment": 0.011342550088712133,
+      "south": -0.008689572376747992,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.0264751529168315,
+      "step_cost": -0.019999999999990636,
+      "complete": 0.08825050972277168
+    },
+    "n_sheep": 3
+  },
+  {
+    "sr": 0.5666666666666667,
+    "mean_len": 2571.866666666667,
+    "mean_min_pen": 2.1761705835660297,
+    "mean_act": 1.2794624905502197,
+    "failure_modes": {
+      "PARTIAL_2of4": 5,
+      "SUCCESS": 17,
+      "DROVE_NO_SHEEP": 1,
+      "NEVER_COMPACT": 1,
+      "PARTIAL_1of4": 6
+    },
+    "reward_per_step": {
+      "progress": 0.04547638401556759,
+      "alignment": 0.003989776116242459,
+      "south": -0.04544084245355691,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.010887034060863705,
+      "step_cost": -0.01999999999998557,
+      "complete": 0.02203328321841464
+    },
+    "n_sheep": 4
+  },
+  {
+    "sr": 0.0,
+    "mean_len": 4000.0,
+    "mean_min_pen": 1.7023075381914774,
+    "mean_act": 1.3590981605617019,
+    "failure_modes": {
+      "PARTIAL_1of5": 9,
+      "PARTIAL_3of5": 2,
+      "PARTIAL_2of5": 2,
+      "PARTIAL_4of5": 17
+    },
+    "reward_per_step": {
+      "progress": 0.039584031492471694,
+      "alignment": 0.003391631218188155,
+      "south": -0.03930825256315925,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.00725,
+      "step_cost": -0.01999999999998423,
+      "complete": 0.0
+    },
+    "n_sheep": 5
+  },
+  {
+    "sr": 0.36666666666666664,
+    "mean_len": 3136.766666666667,
+    "mean_min_pen": 1.7896055857340494,
+    "mean_act": 1.3694271957435262,
+    "failure_modes": {
+      "SUCCESS": 11,
+      "PARTIAL_3of6": 5,
+      "PARTIAL_4of6": 14
+    },
+    "reward_per_step": {
+      "progress": 0.06539200159542725,
+      "alignment": 0.00849681660918308,
+      "south": -0.03917853851677538,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.014558515669000988,
+      "step_cost": -0.019999999999984894,
+      "complete": 0.011689319150292764
+    },
+    "n_sheep": 6
+  },
+  {
+    "sr": 0.0,
+    "mean_len": 4000.0,
+    "mean_min_pen": 1.8426543315251669,
+    "mean_act": 1.383490810132896,
+    "failure_modes": {
+      "PARTIAL_5of7": 18,
+      "PARTIAL_3of7": 3,
+      "PARTIAL_6of7": 7,
+      "PARTIAL_4of7": 2
+    },
+    "reward_per_step": {
+      "progress": 0.05331589305400848,
+      "alignment": 0.00686667034524816,
+      "south": -0.03559404498259062,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.012416666666666666,
+      "step_cost": -0.01999999999998423,
+      "complete": 0.0
+    },
+    "n_sheep": 7
+  },
+  {
+    "sr": 0.6,
+    "mean_len": 2472.0666666666666,
+    "mean_min_pen": 1.609976100921631,
+    "mean_act": 1.3901071324053385,
+    "failure_modes": {
+      "SUCCESS": 18,
+      "PARTIAL_4of8": 3,
+      "PARTIAL_6of8": 9
+    },
+    "reward_per_step": {
+      "progress": 0.09555249268374479,
+      "alignment": 0.010622170754243947,
+      "south": -0.050827120587952045,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.028316388446913515,
+      "step_cost": -0.019999999999985724,
+      "complete": 0.024271190097354442
+    },
+    "n_sheep": 8
+  },
+  {
+    "sr": 0.0,
+    "mean_len": 4000.0,
+    "mean_min_pen": 1.5165573159853618,
+    "mean_act": 1.3936255563423037,
+    "failure_modes": {
+      "PARTIAL_8of9": 26,
+      "PARTIAL_7of9": 4
+    },
+    "reward_per_step": {
+      "progress": 0.07870613823334376,
+      "alignment": 0.007913931652916581,
+      "south": -0.01837508026464782,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.019666666666666666,
+      "step_cost": -0.01999999999998423,
+      "complete": 0.0
+    },
+    "n_sheep": 9
+  },
+  {
+    "sr": 0.8333333333333334,
+    "mean_len": 2243.0,
+    "mean_min_pen": 1.5175361116727193,
+    "mean_act": 1.3979439154633806,
+    "failure_modes": {
+      "SUCCESS": 25,
+      "PARTIAL_7of10": 2,
+      "PARTIAL_8of10": 3
+    },
+    "reward_per_step": {
+      "progress": 0.13872242361851903,
+      "alignment": 0.015031396471697371,
+      "south": -0.04365405077614671,
+      "compact": 0.0,
+      "wall_touch": 0.0,
+      "pen_bonus": 0.04279982166740972,
+      "step_cost": -0.019999999999986123,
+      "complete": 0.03715262297518205
+    },
+    "n_sheep": 10
+  }
+]