A model experiment must be reproducible, and the team must be able to identify which data and implementation produced its metrics. Which two records are necessary? Select 2 answers.

1. Only the model author’s final presentation screenshot.
2. The training code, configuration, and artifact versions used by the run.
3. A mutable dataset name without a recorded snapshot.
4. An overall metric value with no definition of the evaluation procedure.
5. A stable version or snapshot of the training and evaluation datasets.