tinker_cookbook.rl.Transition

class tinker_cookbook.rl.Transition()

A single (observation, action, reward) tuple from a trajectory.

Fields:

• ob ( Observation) – Observation the agent saw before taking the action.
• ac ( TokensWithLogprobs) – Action taken (tokens and their log-probabilities).
• reward ( float) – Immediate reward received after taking the action.
• episode_done ( bool) – Whether this transition ended the episode.
• metrics ( Metrics, default: field(default_factory=dict)) – Numeric values aggregated and reported in training logs.
• logs ( Logs, default: field(default_factory=dict)) – Diagnostic info for display/debugging tools (not aggregated like metrics).

Referenced by

• tinker_cookbook.rl.Trajectory