Struct ReplayBuffer 

pub struct ReplayBuffer { /* private fields */ }

Fixed-capacity FIFO replay buffer for off-policy training.

Stores (obs, action, reward, next_obs, done) transitions and supports uniform random sampling via sampling::sample. The buffer is allocated once at construction (vec![0.0; capacity * obs_dim] etc.); no further allocations occur during push.

Implementations

impl ReplayBuffer

pub fn new(capacity: usize, obs_dim: usize) -> Self

Create a new replay buffer with the given capacity and observation dimensionality.

Arguments

• capacity - Maximum number of transitions stored. Must be > 0.
• obs_dim - Length of one observation vector. Must be > 0.

Panics

Panics if capacity == 0 or obs_dim == 0.

pub fn push( &mut self, obs: &[f32], action: i64, reward: f32, next_obs: &[f32], done: bool, )

Push a single transition into the buffer.

When the buffer is at capacity, this overwrites the oldest transition (FIFO eviction).

Arguments

• obs - Current observation; must have length obs_dim.
• action - Discrete action taken.
• reward - Reward received.
• next_obs - Resulting observation; must have length obs_dim.
• done - true if the episode terminated (used to mask the bootstrap term in the TD target).

Panics

Panics in debug builds if obs.len() != obs_dim or next_obs.len() != obs_dim.

pub fn len(&self) -> usize

Number of transitions currently in the buffer.

pub fn is_empty(&self) -> bool

true if the buffer holds no transitions.

pub fn capacity(&self) -> usize

Maximum number of transitions the buffer can hold.

pub fn obs_dim(&self) -> usize

Observation dimension (length of one obs slice).

pub fn is_ready(&self, min_size: usize) -> bool

true if the buffer holds at least min_size transitions.

Used by the trainer to decide whether to start sampling / performing gradient updates: vanilla DQN normally collects a few thousand transitions before the first update.

Trait Implementations

impl Clone for ReplayBuffer

fn clone(&self) -> ReplayBuffer

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ReplayBuffer

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.