Skip to content

Latest commit

 

History

History
9 lines (7 loc) · 438 Bytes

README.md

File metadata and controls

9 lines (7 loc) · 438 Bytes

dcyphr NLP playground

Main Idea

Fine-tune CNN-DM pretrained BART on dcyphr dataset of manually produced summaries of biomedical literature. Deploy model in production so that dcyphr users have access to automatic summarization and can reinforce the model by manually pruning the output.

Todo

  • Fine-tune pretrained BART on dcyphr data
  • Discover alternative datasets for fine-tuning task
  • Build out human in the loop pipeline