tju-drl-lab/ai-optimizer

The next generation deep reinforcement learning tookit

ai-optimizer's Issues

Specify Python version in the Readme.md

I am not sure whether the python version matters in deep reinforcement learning. As I know, probably python 3.7 is the most stable and good used version for machine learning. Could you please specify what version you guys used and add this information to the Readme file? Many thanks.

Is the sampled muzero Implementation complete?

I compared the sampled muzero code with the muzero general but I didn't find the code about the number of samples and the policy improvement, can you tell me what changes you have made?

why i couldn't run the program MAIN_DDPG.PY

D:\Anaconda3\envs\marl\python.exe "D:\Program Files\JetBrains\PyCharm 2021.1.3\plugins\python\helpers\pycharm_jb_pytest_runner.py" --path "E:/CODE GIT/AI-Optimizer/multiagent-rl/easy-marl/main_ddpg.py"
Testing started at 14:55 ...
D:\Program Files\JetBrains\PyCharm 2021.1.3\plugins\python\helpers\pycharm_jb_pytest_runner.py:28: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
Launching pytest with arguments E:/CODE GIT/AI-Optimizer/multiagent-rl/easy-marl/main_ddpg.py --no-header --no-summary -q in E:\CODE GIT\AI-Optimizer\multiagent-rl\easy-marl

if version.LooseVersion(pytest.version) >= version.LooseVersion("6.0"):
============================= test session starts =============================
collecting ... collected 1 item

main_ddpg.py::test_one_episode ERROR [100%]
**test setup failed
file E:\CODE GIT\AI-Optimizer\multiagent-rl\easy-marl\main_ddpg.py, line 7
def test_one_episode(args, agent):
E fixture 'args' not found

  available fixtures: cache, capfd, capfdbinary, caplog, capsys, capsysbinary, doctest_namespace, monkeypatch, pytestconfig, record_property, record_testsuite_property, record_xml_attribute, recwarn, tmp_path, tmp_path_factory, tmpdir, tmpdir_factory
  use 'pytest --fixtures [testpath]' for help on them.**

E:\CODE GIT\AI-Optimizer\multiagent-rl\easy-marl\main_ddpg.py:7

============================== 1 error in 0.86s ===============================

进程已结束，退出代码为 1

MAPPO的critic为什么不是一个，而是多个呢？