search

timoschick / pet

This repository contains the code for "Exploiting Cloze Questions for Few-Shot Text Classification and Natural Language Inference"
https://arxiv.org/abs/2001.07676
Apache License 2.0
1.61k stars 285 forks source link

Seed when used in training? #59

Open sherlcok314159 opened 2 years ago

sherlcok314159 commented 2 years ago

I am trying to reproduce your work. While in your paper, I only know that you use three seeds and get average of them. But could you tell me what exactly seeds you use in training?

Regards, Yunpeng Tai

timoschick commented 2 years ago

Hi @sherlcok314159, could you specify which experiments exactly you want to reproduce (i.e., which paper and which datasets are you referring to)?

BohanLi0110 commented 2 years ago

Hi @timoschick , I'm now trying to reproduce your work, too. Could you kindly provide all the seeds you use on PET in these two papers? lol

That is, Yelp, AG, Yahoo, and MNLI in Exploiting Cloze Questions for Few-Shot Text Classification and Natural Language Inference; and 8 datasets from FewGlue in It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners

Cheers!