TensorUI / relative-position-pytorch

a pytorch implementation of self-attention with relative position representations
51 stars 17 forks source link

Can you give some descriptions and explanations of your code? #3

Open smiles724 opened 3 years ago

smiles724 commented 3 years ago

Thanks for your implementation of RPE. Recently I am just trying to do this. However, do you mind add some explanations in the readme file?

Rabailkamboh commented 1 year ago

hey. did u understand it? how to implement it in transformer original model?