getting MlpLstmPolicy working with PPO DQN or A2C??

[chrisantha]

I'm not sure how to get the LSTM policy working with the above algorithms. It says

model = PPO('MlpLstmPolicy', env, verbose=1).learn(5000)

File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/stable_baselines3/ppo/ppo.py", line 110, in init
_init_setup_model=False,
File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/stable_baselines3/common/on_policy_algorithm.py", line 87, in init
tensorboard_log=tensorboard_log,
File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/stable_baselines3/common/base_class.py", line 106, in init
self.policy_class = get_policy_from_name(policy_base, policy)
File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/stable_baselines3/common/policies.py", line 815, in get_policy_from_name
f"Error: unknown policy type {name},"
KeyError: "Error: unknown policy type MlpLstmPolicy,the only registed policy type are: ['MlpPolicy', 'CnnPolicy']!"

[Miffyli]

There is no support for LSTM/RNN policies in stable-baselines3 yet. You need to use stable-baselines for those.

Closing as duplicate of #18.