Struct border_atari_env::BorderAtariEnv

pub struct BorderAtariEnv<O, A, OF, AF>where
    O: Obs,
    A: Act,
    OF: BorderAtariObsFilter<O>,
    AF: BorderAtariActFilter<A>,{ /* private fields */ }

A wrapper of atari learning environment.

Preprocessing is the same in the link: https://stable-baselines3.readthedocs.io/en/master/common/atari_wrappers.html#stable_baselines3.common.atari_wrappers.AtariWrapper.

Implementations

impl<O, A, OF, AF> BorderAtariEnv<O, A, OF, AF>where O: Obs, A: Act, OF: BorderAtariObsFilter<O>, AF: BorderAtariActFilter<A>,

pub fn open(&mut self) -> Result<()>

Opens window for display.

pub fn get_num_actions_atari(&self) -> i64

Returns the number of actions.

Trait Implementations

impl<R: ReplayBufferBase> Agent<BorderAtariEnv<BorderAtariObs, BorderAtariAct, BorderAtariObsRawFilter<BorderAtariObs>, BorderAtariActRawFilter<BorderAtariAct>>, R> for RandomAgent

fn train(&mut self)

Set the policy to training mode.

fn eval(&mut self)

Set the policy to evaluation mode.

fn is_train(&self) -> bool

Return if it is in training mode.

fn opt(&mut self, buffer: &mut R) -> Option<Record>

Do an optimization step.

fn save<T: AsRef<Path>>(&self, _path: T) -> Result<()>

Save the agent in the given directory. This method commonly creates a number of files consisting the agent in the directory. For example, the DQN agent in border_tch_agent crate saves two Q-networks corresponding to the original and target networks.

fn load<T: AsRef<Path>>(&mut self, _path: T) -> Result<()>

Load the agent from the given directory.

impl<O, A, OF, AF> Default for BorderAtariEnv<O, A, OF, AF>where O: Obs, A: Act, OF: BorderAtariObsFilter<O>, AF: BorderAtariActFilter<A>,

fn default() -> Self

Returns the “default value” for a type.

impl<O, A, OF, AF> Env for BorderAtariEnv<O, A, OF, AF>where O: Obs, A: Act, OF: BorderAtariObsFilter<O>, AF: BorderAtariActFilter<A>,

fn step_with_reset(&mut self, a: &Self::Act) -> (Step<Self>, Record)where Self: Sized,

Currently it supports non-vectorized environment.

type Config = BorderAtariEnvConfig<O, A, OF, AF>

Configurations.

type Obs = O

Observation of the environment.

type Act = A

Action of the environment.

type Info = NullInfo

Information in the self::Step object.

fn build(config: &Self::Config, _seed: i64) -> Result<Self>where Self: Sized,

Builds an environment with a given random seed.

fn reset(&mut self, _is_done: Option<&Vec<i8>>) -> Result<Self::Obs>

Resets the environment if is_done[0] == 1. or is_done.is_none().

fn reset_with_index(&mut self, ix: usize) -> Result<Self::Obs>

Resets the environment with a given index.

fn step(&mut self, act: &Self::Act) -> (Step<Self>, Record)where Self: Sized,

Performes an environment step.

impl Policy<BorderAtariEnv<BorderAtariObs, BorderAtariAct, BorderAtariObsRawFilter<BorderAtariObs>, BorderAtariActRawFilter<BorderAtariAct>>> for RandomAgent

type Config = RandomAgentConfig

Configuration of the policy.

fn build(config: Self::Config) -> Self

Builds the policy.

fn sample(&mut self, _: &Obs) -> Act

Sample an action given an observation.