# tinker_cookbook.rl.assemble_training_data

### tinker_cookbook.rl.assemble_training_data( _trajectory_groups_P_, _advantages_P_)

Convert trajectories to training data format.

**Parameters:**

- **trajectory_groups_P** ( _list[ [TrajectoryGroup](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/rl/trajectorygroup/) ]_) – Groups of trajectories to convert into training datums.
- **advantages_P** ( _list[torch.Tensor]_) – Per-group advantage tensors, one per trajectory group, as returned by [`compute_advantages`](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/rl/compute_advantages/).

**Returns:** _tuple[list[tinker. [Datum](https://tinker-docs.thinkingmachines.ai/tinker/api-reference/types/datum/) ], list[dict[str, int]]]_ – A flat list of training datums and a parallel list of metadata dicts mapping each datum back to its `group_idx` and `traj_idx`.