Resolve this error about finrl-tutorials HOT 7 CLOSED

it need to change to : agent.states = env.reset()[0][np.newaxis, :]

from finrl-tutorials.

Thank you

from finrl-tutorials.

in explore_env(self, env, horizon_len)
201
202 ary_action = convert(action).detach().cpu().numpy()
--> 203 ary_state, reward, done, _ = env.step(ary_action)
204 if done:
205 ary_state = env.reset()

ValueError: too many values to unpack (expected 4). Now it is showing another error.

from finrl-tutorials.

all of the env.reset() you need to do the same env.reset()[0]

from finrl-tutorials.

but I am still fighting to understand how everything is done, wish they had a manual.... it is sooo cryptic.... I been trying to modify the models, but they dont improve the learning....

from finrl-tutorials.

Yes, model is showing error in every step. Thanks a lot for your help.

from finrl-tutorials.

After modifying env.reset() to env.reset()[0] this error is resolved.

from finrl-tutorials.