文件名称:dqn_graphs:使用Deep Q Networks和相关算法(DDQN,对决DDQN)找到极值ex(n,H)
文件大小:10.71MB
文件格式:ZIP
更新时间:2024-04-20 14:30:26
Python
dqn_graphs 使用Deep Q Networks和相关算法(DDQN,对决DDQN)找到极值ex(n,H) DQN,DDQN等算法的代码直接来自 看看他在该主题上的出色课程! 第一个版本。 您需要在c4free_env.py文件中更改N,我需要弄清楚如何从那里获取它。 通过进入主文件夹并编写pip install -e gym-c4free来安装环境。 通过python main.py -n_games 100 -algo DuelingDDQNAgent -path models /运行 确保主文件夹中有一个空的“模型”和“图形”文件夹,否则脚本将在最后崩溃
【文件预览】:
dqn_graphs-main
----models()
--------gym_c4free-c4free-v0_DuelingDDQNAgent_q_eval(2.42MB)
--------gym_c4free-c4free-v0_DuelingDDQNAgent_q_next(2.42MB)
----models_test()
--------gym_c4free-c4free-v0_DuelingDDQNAgent_q_eval(3.23MB)
--------gym_c4free-c4free-v0_DuelingDDQNAgent_q_next(3.23MB)
----utils.py(1KB)
----agents.py(8KB)
----main.py(4KB)
----c4count_rewardtest.py(913B)
----plots()
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.001_1000games0.99987epsdec.png(35KB)
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.0001_2000games.png(29KB)
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.001_1000games.png(40KB)
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.0001_1000games0.99987epsdec.png(32KB)
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.0001_600games.png(42KB)
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.0001_60000games.png(25KB)
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.001_3000games0.99987epsdec.png(29KB)
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.001_10000games.png(42KB)
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.0001_120games.png(37KB)
--------DuelingDDQNAgent_gym_c4free-c4free-v0_alpha0.01_2000games0.9999epsdec.png(36KB)
----README.md(774B)
----gym-c4free()
--------gym_c4free()
--------gym_c4free.egg-info()
--------setup.py(155B)
----__pycache__()
--------replay_memory.cpython-36.pyc(1KB)
--------deep_q_network.cpython-36.pyc(4KB)
--------utils.cpython-36.pyc(1KB)
--------agents.cpython-36.pyc(8KB)
----replay_memory.py(1KB)
----deep_q_network.py(3KB)