search

harvardnlp / encoder-agnostic-adaptation

Encoder-Agnostic Adaptation for Conditional Language Generation
https://arxiv.org/abs/1908.06938
MIT License
79 stars 13 forks source link

Image captioning with GPT-2 1.5B #7

Open ftarlaci opened 4 years ago

ftarlaci commented 4 years ago

Are you planning on using the biggest GPT-2 for the image captioning model you are working on?

lukemelas commented 4 years ago

Thanks for the question! We're not planning on using the 1.5B model for captioning at the moment, as we're working on other things (and there are non-trivial compute constraints when using the 1.5B model). That being said, you're welcome to try it out!