Struct PrioritizedBatch 

pub struct PrioritizedBatch {
    pub observations: Vec<f32>,
    pub actions: Vec<i64>,
    pub rewards: Vec<f32>,
    pub next_observations: Vec<f32>,
    pub dones: Vec<bool>,
    pub is_weights: Vec<f32>,
    pub indices: Vec<usize>,
    pub obs_dim: usize,
}

One minibatch sampled from a PrioritizedReplayBuffer.

All fields are CPU-side primitive vectors; convert to Burn tensors via PrioritizedBatch::to_burn_tensors when handing them to the trainer.

Fields

observations: Vec<f32>

Flattened current observations, shape [batch_size * obs_dim].

actions: Vec<i64>

Actions taken, length batch_size.

rewards: Vec<f32>

Rewards received, length batch_size.

next_observations: Vec<f32>

Flattened next observations, shape [batch_size * obs_dim].

dones: Vec<bool>

Episode-end mask, length batch_size.

is_weights: Vec<f32>

Per-sample importance-sampling weights, length batch_size, normalized so max(weights) == 1.0.

indices: Vec<usize>

Sum-tree leaf indices for the sampled transitions, length batch_size. Pass these back into PrioritizedReplayBuffer::update_priorities alongside the new TD errors.

obs_dim: usize

Length of one observation slice.

Implementations

impl PrioritizedBatch

pub fn len(&self) -> usize

Number of transitions in the batch.

pub fn is_empty(&self) -> bool

true if the batch is empty.

pub fn to_burn_tensors<B: Backend>( &self, device: &B::Device, ) -> PrioritizedBurnTensors<B>

Stack the batch into Burn tensors on device.

Returns a named PrioritizedBurnTensors struct so the DQN trainer can grab fields by name. Includes the importance-sampling weights that the uniform super::ReplayBatch does not carry.

Shapes (all on device):

• observations: [batch, obs_dim], f32
• actions: [batch], i64
• rewards: [batch], f32
• next_observations: [batch, obs_dim], f32
• dones: [batch], f32 (0.0 / 1.0)
• is_weights: [batch], f32

Note: indices is not a tensor — leaf-index round-trips back into PrioritizedReplayBuffer::update_priorities as a host &[usize], so it stays on self.

Trait Implementations

impl Clone for PrioritizedBatch

fn clone(&self) -> PrioritizedBatch

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PrioritizedBatch

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

Formats the value using the given formatter.