[dev] update

README.md

@@ -66,6 +66,7 @@ class EnvCore(object):
 
 只需要编写这一部分的代码，就可以无缝衔接MAPPO。在env_core.py之后，单独提出来了两个文件env_discrete.py和env_continuous.py这两个文件用于封装处理动作空间和离散动作空间。在algorithms/utils/act.py中elif self.continuous_action:这个判断逻辑也是用来处理连续动作空间的。和runner/shared/env_runner.py部分的# TODO 这里改造成自己环境需要的形式即可都是用来处理连续动作空间的。
 
+在train.py文件里面，选择注释连续环境，或者离散环境进行demo环境的切换。
 
 ## Related Efforts
 

envs/env_discrete.py

@@ -82,7 +82,7 @@ class DiscreteActionEnv(object):
     def seed(self, seed):
         pass
 
-class MultiDiscrete(gym.Space):
+class MultiDiscrete():
     """
     - The multi-discrete action space consists of a series of discrete action spaces with different parameters
     - It can be adapted to both a Discrete action space or a continuous (Box) action space

runner/separated/env_runner.py

@@ -133,7 +133,9 @@ class EnvRunner(Runner):
             elif self.envs.action_space[agent_id].__class__.__name__ == 'Discrete':
                 action_env = np.squeeze(np.eye(self.envs.action_space[agent_id].n)[action], 1)
             else:
-                raise NotImplementedError
+                # TODO 这里改造成自己环境需要的形式即可
+                action_env = actions
+                # raise NotImplementedError
 
             actions.append(action)
             temp_actions_env.append(action_env)

runner/shared/env_runner.py

@@ -5,18 +5,11 @@
 # @File    : env_runner.py
 """
 
-"""
-# @Time    : 2021/7/1 7:04 下午
-# @Author  : hezhiqiang01
-# @Email   : hezhiqiang01@baidu.com
-# @File    : huaru_runner.py
-"""
-
 import time
 import numpy as np
 import torch
 from runner.shared.base_runner import Runner
-import imageio
+# import imageio
 
 
 def _t2n(x):
@@ -269,5 +262,5 @@ class EnvRunner(Runner):
 
             print("average episode rewards is: " + str(np.mean(np.sum(np.array(episode_rewards), axis=0))))
 
-        if self.all_args.save_gifs:
-            imageio.mimsave(str(self.gif_dir) + '/render.gif', all_frames, duration=self.all_args.ifi)
+        # if self.all_args.save_gifs:
+        #     imageio.mimsave(str(self.gif_dir) + '/render.gif', all_frames, duration=self.all_args.ifi)

train/train.py

@@ -22,10 +22,11 @@ from envs.env_wrappers import SubprocVecEnv, DummyVecEnv
 def make_train_env(all_args):
     def get_env_fn(rank):
         def init_env():
-            # from envs.env_continuous import ContinuousActionEnv
-            # env = ContinuousActionEnv()
-            from envs.env_discrete import DiscreteActionEnv
-            env = DiscreteActionEnv()
+            # TODO 注意注意，这里选择连续还是离散可以选择注释上面两行，或者下面两行。
+            from envs.env_continuous import ContinuousActionEnv
+            env = ContinuousActionEnv()
+            # from envs.env_discrete import DiscreteActionEnv
+            # env = DiscreteActionEnv()
             env.seed(all_args.seed + rank * 1000)
             return env
         return init_env
@@ -35,8 +36,11 @@ def make_train_env(all_args):
 def make_eval_env(all_args):
     def get_env_fn(rank):
         def init_env():
-            from envs.env_discrete import DiscreteActionEnv
-            env = DiscreteActionEnv()
+            # TODO 注意注意，这里选择连续还是离散可以选择注释上面两行，或者下面两行。
+            from envs.env_continuous import ContinuousActionEnv
+            env = ContinuousActionEnv()
+            # from envs.env_discrete import DiscreteActionEnv
+            # env = DiscreteActionEnv()
             env.seed(all_args.seed + rank * 1000)
             return env
         return init_env