Howto RL-AGENT-001: Run an Agent with Own Policy
Prerequisites
Please install the following packages to run this examples properly:
Executable code
## -------------------------------------------------------------------------------------------------
## -- Project : MLPro - A Synoptic Framework for Standardized Machine Learning Tasks
## -- Package : mlpro.rl.examples
## -- Module : howto_rl_agent_001_run_agent_with_own_policy_on_gym_environment.py
## -------------------------------------------------------------------------------------------------
## -- History :
## -- yyyy-mm-dd Ver. Auth. Description
## -- 2021-05-09 0.0.0 DA Creation
## -- 2021-06-06 1.0.0 DA Released first version
## -- 2021-08-28 1.1.0 DA Introduced Policy
## -- 2021-09-11 1.1.0 MRD Change Header information to match our new library name
## -- 2021-09-29 1.1.1 SY Change name: WrEnvGym to WrEnvGYM2MLPro
## -- 2021-10-06 1.1.2 DA Refactoring
## -- 2021-10-18 1.1.3 DA Refactoring
## -- 2021-11-15 1.2.0 DA Refactoring
## -- 2021-11-16 1.2.1 DA Added explicit scenario reset with constant seeding
## -- 2021-12-03 1.2.2 DA Refactoring
## -- 2022-07-20 1.2.3 SY Update due to the latest introduction of Gym 0.25
## -- 2022-10-13 1.2.4 SY Refactoring
## -- 2022-11-01 1.2.5 DA Refactoring
## -- 2022-11-02 1.2.6 DA Refactoring
## -- 2022-11-07 1.3.0 DA Refactoring
## -- 2023-01-14 1.3.1 MRD Removing default parameter new_step_api and render_mode for gym
## -- 2023-04-19 1.3.2 MRD Refactor module import gym to gymnasium
## -------------------------------------------------------------------------------------------------
"""
Ver. 1.3.2 (2023-04-19)
This module shows how to run an own policy inside the standard agent model with an OpenAI Gym environment using
MLPro framework.
You will learn:
1) How to set up a native policy for an agent
2) How to set up an agent
3) How to set up a scenario
4) How to wrap Gym environment to MLPro environment
5) How to run the scenario
"""
from mlpro.bf.math import *
from mlpro.rl import *
from mlpro.wrappers.gymnasium import WrEnvGYM2MLPro
import gymnasium as gym
import random
# 1 Implement your own agent policy
class MyPolicy (Policy):
C_NAME = 'MyPolicy'
def set_random_seed(self, p_seed=None):
random.seed(p_seed)
def compute_action(self, p_state: State) -> Action:
# 1.1 Create a numpy array for your action values
my_action_values = np.zeros(self._action_space.get_num_dim())
# 1.2 Computing action values is up to you...
for d in range(self._action_space.get_num_dim()):
my_action_values[d] = random.random()
# 1.3 Return an action object with your values
return Action(self._id, self._action_space, my_action_values)
def _adapt(self, p_sars_elem: SARSElement) -> bool:
# 1.4 Adapting the internal policy is up to you...
self.log(self.C_LOG_TYPE_W, 'Sorry, I am a stupid agent...')
# 1.5 Only return True if something has been adapted...
return False
# 2 Implement your own RL scenario
class MyScenario (RLScenario):
C_NAME = 'Matrix'
def _setup(self, p_mode, p_ada: bool, p_visualize:bool, p_logging) -> Model:
# 2.1 Setup environment
if p_visualize:
gym_env = gym.make('CartPole-v1', render_mode="human")
else:
gym_env = gym.make('CartPole-v1')
self._env = WrEnvGYM2MLPro( p_gym_env=gym_env, p_visualize=p_visualize, p_logging=p_logging)
# 2.2 Setup standard single-agent with own policy
return Agent( p_policy=MyPolicy( p_observation_space=self._env.get_state_space(),
p_action_space=self._env.get_action_space(),
p_buffer_size=1,
p_ada=p_ada,
p_visualize=p_visualize,
p_logging=p_logging),
p_envmodel=None,
p_name='Smith',
p_ada=p_ada,
p_visualize=p_visualize,
p_logging=p_logging)
# 3 Create scenario and run some cycles
if __name__ == "__main__":
# 3.1 Parameters for demo mode
cycle_limit = 100
logging = Log.C_LOG_ALL
visualize = True
else:
# 3.2 Parameters for internal unit test
cycle_limit = 10
logging = Log.C_LOG_NOTHING
visualize = False
# 3.3 Create your scenario and run some cycles
myscenario = MyScenario(
p_mode=Mode.C_MODE_SIM,
p_ada=True,
p_cycle_limit=cycle_limit,
p_visualize=visualize,
p_logging=logging
)
myscenario.reset(p_seed=3)
myscenario.run()
Results
The OpenAI Gym cartpole window appears and shows a random behavior while the demo run is logged on the console. Demo should terminate after 61 cycles because of the fixed random seed.
Cross Reference