1 rollouts.replay.types

rollouts.replay.types

Minimal score contract consumed by finite-candidate replay.

This module provides the binding from compact evaluator scores to the full-shell hard mask and stable shell indices required for action selection.

1.1 Classes

Name	Description
CandidateScores	Scores aligned to the hard-valid rows of one candidate table.

1.1.1 CandidateScores

rollouts.replay.types.CandidateScores(
    values,
    action_mask,
    candidate_shell_indices,
    name,
)

Scores aligned to the hard-valid rows of one candidate table.

1.1.1.1 Attributes

Name	Description
values	Compact score vector in stable valid-candidate order.
action_mask	Full-shell hard action mask; invalid rows are never selectable.
candidate_shell_indices	Full-shell row ids corresponding one-to-one with values.
name	Stable semantic name of the score used for action selection.

1.1.1.2 Methods

Name	Description
from_valid_values	Normalize compact scores and bind them to one hard-masked table.
validate_for	Verify the hard mask and candidate-order link against candidates.

1.1.1.2.1 from_valid_values

rollouts.replay.types.CandidateScores.from_valid_values(
    values,
    *,
    name,
    candidates,
    device,
    dtype,
)

Normalize compact scores and bind them to one hard-masked table.

1.1.1.2.2 validate_for

rollouts.replay.types.CandidateScores.validate_for(candidates, *, device, dtype)

Verify the hard mask and candidate-order link against candidates.