LearnerBatch
python
LearnerBatch(states: torch.Tensor, actions: torch.Tensor, rewards: torch.Tensor, next_states: torch.Tensor, dones: torch.Tensor, n_steps: torch.Tensor, weights: torch.Tensor | None = None) -> NoneLogical batch of training tensors used by the learner.
Values
actions
python
actions: torch.Tensordones
python
dones: torch.Tensorn_steps
python
n_steps: torch.Tensornext_states
python
next_states: torch.Tensorrewards
python
rewards: torch.Tensorstates
python
states: torch.Tensorweights
python
weights: torch.Tensor | None = None