Module `Openai_gym.Gym_client`

val base_url : string Stdlib.ref: Server url: default is http://127.0.0.1:5000

val env_create : string -> Gym_t.instance_id: env_create env_id creates an instance of the specified environment (e.g., "CartPole-v0"). It returns the instance identifier.

val env_list_all : unit -> (Gym_t.instance_id * string) list: env_list_all () lists all the environments running on the server as a pair (instance_id, env_id) (e.g. [("3c657dbc", "CartPole-v0")]).

val env_reset : Gym_t.instance_id -> Gym_t.observation: env_reset instance_id resets the state of the environment and return an initial observation.

val env_step : Gym_t.instance_id -> Gym_t.action -> bool -> Gym_t.step_response: env_step instance_id action render steps though an environment using an action. If render is true, a graphical feedback if display by the server.

val env_action_space_info : Gym_t.instance_id -> Gym_t.action_space_info: env_action_space_info instance_id gets information (name and dimensions/bounds) of the env's action_space.

val env_action_space_sample : Gym_t.instance_id -> Gym_t.action: env_action_space_sample instance_id samples randomly from the env's action_space.

val env_action_space_contains : Gym_t.instance_id -> int -> bool: env_action_space_contains instance_id x checks to see if the value x is valid in the env's action_space.

val env_observation_space_info : Gym_t.instance_id -> Gym_t.observation_space_info: env_observation_space_info instance_id gets information (name and dimensions/bounds) of the env's observation_space.

val env_observation_space_contains : Gym_t.instance_id -> Json_t.json -> bool: env_observation_space_contains instance_id params assesses that the parameters are members of the env's observation_space.

val env_monitor_start : Gym_t.instance_id -> string -> bool -> bool -> unit: env_monitor_start instance_id directory force resume starts monitoring. force clears out existing training data from this directory (by deleting every file prefixed with openaigym.). resume retains the training data already in this directory, which will be merged with our new data.

val env_monitor_close : Gym_t.instance_id -> unit: env_monitor_close instance_id flushes all monitor data to disk.

val env_close : Gym_t.instance_id -> unit: env_close instance_id stops the environment.

val shutdown_server : unit -> string: shutdown_server () requests a server shutdown.