search

AI4Finance-Foundation / ElegantRL

Massively Parallel Deep Reinforcement Learning. 🔥
https://ai4finance.org
Other
3.67k stars 842 forks source link

a typeError in build_env() function #358

Open hululu-lzb opened 4 months ago

hululu-lzb commented 4 months ago

I inherited from gym.Env and created a new class, passing in the additional parameter env_config in the init method of this new class. But when I use train_agent(), the built_env() will report an error, check elegantRL source code, this is because my gym.make is gym.envs.registration. So by default no parameters are passed when creating an environment. Is there any solution? image1 image2

hululu-lzb commented 4 months ago

My demo is copied from fllow： image

sonderli commented 4 months ago

I meet the same quesation,did you solve it?

stellawang196 commented 4 months ago

gym.make() is used to initialize an environment specified by a unique name (ID). This ID corresponds to predefined environment configurations registered in Gym. If the env to make is from the registration system, such as CartPole-v1, the module will be gym.envs.registration.

To use self-defined env class inherited from gym.Env, there is no need to use gym.make().