dojo.policies

Policies are the means to give agent behaviour.

Specifically, policies map observation spaces to actions and are attached to agents

class AAVEv3Policy(dojo.policies.BasePolicy[dojo.actions.aaveV3.BaseAaveAction, dojo.agents.aaveV3.AAVEv3Agent, dojo.observations.aaveV3.AAVEv3Observation]):

A policy for AAVEv3.

class GmxV2Policy(dojo.policies.BasePolicy[dojo.actions.gmxV2.base_gmx_action.BaseGmxAction, dojo.agents.gmxV2.GmxV2Agent, dojo.observations.gmxV2.GmxV2Observation]):

A policy for GMXv2.

class UniswapV3Policy(dojo.policies.BasePolicy[dojo.actions.uniswapV3.BaseUniswapV3Action, dojo.agents.uniswapV3.UniswapV3Agent, dojo.observations.uniswapV3.UniswapV3Observation]):

A policy for UniswapV3.

class BasePolicy(abc.ABC, typing.Generic[~Action, ~Agent, ~Observation]):

The BasePolicy acts as the base abstract class for all policies.

BasePolicy(agent: ~Agent)

Initialize the policy.

Parameters
  • agent: the agent associated with the policy
@abstractmethod
def predict(self, obs: ~Observation) -> list[~Action]:

Run the policy to get a sequence of actions.

Parameters
  • obs: The observation from the environment.
Returns

The policy actions.