Reinforcement Learning

Pytorch/Project/CartPole/TestEnv.py

@@ -0,0 +1,22 @@
+import gymnasium as gym
+import time
+
+# 创建环境
+env = gym.make("CartPole-v1", render_mode="human")  # human模式会用pyglet显示窗口
+
+# 重置环境
+observation, info = env.reset()
+print("初始观察值:", observation)
+print("<UNK>:", info)
+
+# 随机动作走几步
+for step in range(100):
+    observation, reward, terminated, truncated, info = env.step(1)
+    print(f"Step {step+1}: observation={observation}, reward={reward}, done={terminated}")
+
+    if terminated or truncated:
+        observation, info = env.reset()
+
+time.sleep(1)
+# 关闭环境
+env.close()