Trait renforce::trainer::EpisodicTrainer

pub trait EpisodicTrainer<S: Space, A: Space, T: Agent<S, A>> {
    fn train_step(&mut self,
                  agent: &mut T,
                  env: &mut Environment<State=S, Action=A>);
    fn train(&mut self, agent: &mut T, env: &mut Environment<State=S, Action=A>);
}

Trains agents 1 "episode" at a time

Required Methods

fn train_step(&mut self,
              agent: &mut T,
              env: &mut Environment<State=S, Action=A>)

Trains agent using 1 "episodes" worth of exploration

fn train(&mut self, agent: &mut T, env: &mut Environment<State=S, Action=A>)

Trains agent to perform well in the environment, potentially acting out multiple episodes

Implementors

• impl<F: Float, S: Space, A: Space, T> EpisodicTrainer<S, A, T> for CrossEntropy<F> where T: Agent<S, A> + ParameterizedFunc<F>
• impl<F: Float, S: Space, A: FiniteSpace, G, T> EpisodicTrainer<S, A, T> for PolicyGradient<F, G> where T: Agent<S, A> + LogDiffFunc<S, A, F>,
        G: GradientDescAlgo<F>