Skip to content

Latest commit

 

History

History
11 lines (8 loc) · 373 Bytes

README.md

File metadata and controls

11 lines (8 loc) · 373 Bytes
Raw

zero-shot

Zero-shot classification of Patient Experience data.

Two working scripts:

  • ht_T_training_Bert.py- fine-tuning a BERT model
  • hg_zero_shot.py- Zero-shot classification

Note- these scripts run pretty heavy processes and can take forever. Make sure to experiment with small samples for the time being, while we explore ways of making them more efficient.