Why training (p-tuning v2) is taking a lot of time compared to regular fine-tuning?

THUDM / P-tuning-v2

An optimized deep prompt tuning strategy comparable to fine-tuning across scales and tasks

Apache License 2.0

1.97k stars 201 forks source link

Closed Sudhandar closed 2 years ago