Original Paper: arXiv
| Code Source | Link | License |
|---|---|---|
| Original Source | GitHub | Apache V2.0 (LICENSE.google.research) |
TODO
TODO
Level 2: In addition to being based on reference code (Level 1), the model in this repository has been studied to provide similar (eyeball validation) loss values when running on provided reference training data.