Open beckybai opened 2 years ago
Ah, my bad - I thought we had included zero-shot in this script. Do you need to reproduce the exact results from our paper? Otherwise, I'd recommend to just use the most recent version of PET (following the suggestions from this issue). While there might be slight differences, results should be very similar to those with the original version.
But this script only supports training from 10, 50 or 100 examples. Is there any script for zero-labeled training examples