dqn跑通一个场景

human_action.py

@@ -13,12 +13,12 @@ print('state:', state)
 # action_series = [1] * 30
 # action_series = [[0.2], [0.4], [0.7], [0.5]]
 # action_series = [[-0.08], [-0.08], [0], [0]]
-action_series = [3, 5, 3, 5, 3, 5, 3, 5, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4,
+action_series = [3, 5, 3, 5, 1, 1, 3, 5, 1, 5, 3, 1, 1, 1, 1, 1, 2, 1, 2, 1, 2, 1,
-                 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4, 3, 4]
+                 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1]
 
 for i in range(100):
     action = action_series[i]
-    state, reward, done, info, _ = env.step(action)
+    state, reward, done, _, info = env.step(action)
     print('state:', state)
     print('reward:', reward)
     if done:

mtkl_sovler2.py

@@ -12,7 +12,7 @@ random.seed(42)
 # ---------------------------
 # 需要修改的超参数
 # ---------------------------
-num_iterations = 300000000
+num_iterations = 10000000000
 # 随机生成分区的行分段数与列分段数
 R = random.randint(0, 3)  # 行分段数
 C = random.randint(0, 3)  # 列分段数

solutions/dqn_params_50_50_3.json

@@ -1,10 +1,10 @@
 {
-    "best_time": 8848.626166217664,
+    "best_time": 8820.015746422654,
     "row_cuts": [
         0,
-        0.2700000000000001,
+        0.2900000000000001,
         0.4700000000000001,
-        0.76,
+        0.77,
         1
     ],
     "col_cuts": [
@@ -26,5 +26,5 @@
         6,
         0
     ],
-    "timestamp": "2025-04-01 20:05:51"
+    "timestamp": "2025-04-03 10:58:44"
 }

visualization.py

@@ -200,7 +200,7 @@ if __name__ == "__main__":
     # 需要修改的超参数
     # ---------------------------
     params_file = 'params_50_50_3'
-    solution_file = r'solutions\finetune_params_50_50_3.json'
+    solution_file = r'solutions\dqn_params_50_50_3_2.json'
 
     with open(params_file + '.yml', 'r', encoding='utf-8') as file:
         params = yaml.safe_load(file)