bacow/easy-rl
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update

Browse Source
This commit is contained in:
JohnJim0816
2021-04-16 14:59:23 +08:00
parent 312b57fdff
commit e4690ac89f
71 changed files with 805 additions and 153 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
codes/DQN/main.py
View File

@@ -5,7 +5,7 @@
@Email: johnjim0816@gmail.com
@Date: 2020-06-12 00:48:57
@LastEditor: John
LastEditTime: 2021-04-04 00:26:47
LastEditTime: 2021-04-13 19:03:39
@Discription:
@Environment: python 3.7.7
'''
@@ -21,15 +21,13 @@ from DQN.agent import DQN
from common.plot import plot_rewards
from common.utils import save_results,make_dir,del_empty_dir
SEQUENCE = datetime.datetime.now().strftime("%Y%m%d-%H%M%S") # obtain current time
SAVED_MODEL_PATH = curr_path+"/saved_model/"+SEQUENCE+'/' # path to save model
RESULT_PATH = curr_path+"/results/"+SEQUENCE+'/' # path to save rewards
make_dir(curr_path+"/saved_model/",curr_path+"/results/")
del_empty_dir(curr_path+"/saved_model/",curr_path+"/results/")
curr_time = datetime.datetime.now().strftime("%Y%m%d-%H%M%S") # obtain current time
class DQNConfig:
def __init__(self):
self.algo = "DQN" # name of algo
self.env = 'CartPole-v0'
self.result_path = curr_path+"/results/" +self.env+'/'+curr_time+'/' # path to save results
self.gamma = 0.95
self.epsilon_start = 1 # e-greedy策略的初始epsilon
self.epsilon_end = 0.01
@@ -37,7 +35,7 @@ class DQNConfig:
self.lr = 0.0001 # learning rate
self.memory_capacity = 10000 # Replay Memory容量
self.batch_size = 32
self.train_eps = 300 # 训练的episode数目
self.train_eps = 10 # 训练的episode数目
self.target_update = 2 # target net的更新频率
self.eval_eps = 20 # 测试的episode数目
self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu") # 检测gpu
@@ -72,14 +70,13 @@ def train(cfg,env,agent):
if __name__ == "__main__":
cfg = DQNConfig()
env = gym.make('CartPole-v0')
env = gym.make(cfg.env)
env.seed(1)
state_dim = env.observation_space.shape[0]
action_dim = env.action_space.n
agent = DQN(state_dim,action_dim,cfg)
rewards,ma_rewards = train(cfg,env,agent)
make_dir(SAVED_MODEL_PATH,RESULT_PATH)
agent.save(path=SAVED_MODEL_PATH)
save_results(rewards,ma_rewards,tag='train',path=RESULT_PATH)
plot_rewards(rewards,ma_rewards,tag="train",algo = cfg.algo,path=RESULT_PATH)
del_empty_dir(SAVED_MODEL_PATH,RESULT_PATH)
make_dir(cfg.result_path)
agent.save(path=cfg.result_path)
save_results(rewards,ma_rewards,tag='train',path=cfg.result_path)
plot_rewards(rewards,ma_rewards,tag="train",algo = cfg.algo,path=cfg.result_path)