25 lines
667 B
Rust
25 lines
667 B
Rust
pub mod actor_learner;
|
|
pub mod algorithms;
|
|
pub mod environment;
|
|
pub mod error;
|
|
pub mod rlhf;
|
|
pub mod standalone_rl;
|
|
pub mod tensor_algorithm_tests;
|
|
|
|
// Core RL types that tests expect
|
|
pub type Action = i32;
|
|
pub type State = Vec<f32>;
|
|
pub type Reward = f32;
|
|
|
|
// Re-export core types
|
|
pub use actor_learner::{ActorLearner, ActorLearnerConfig, ActorMessage, LearnerMessage};
|
|
pub use environment::{Environment, Step};
|
|
pub use error::{RLError, Result};
|
|
pub use standalone_rl::{
|
|
Experience as StandaloneExperience, Matrix, PPO as StandalonePPO,
|
|
ReplayBuffer as StandaloneReplayBuffer,
|
|
};
|
|
|
|
// Re-export Experience for compatibility
|
|
pub use standalone_rl::Experience;
|