search

turboLJY / Few-Shot-KG2Text

Source for the ACL 2021 Findings paper "Few-shot Knowledge Graph-to-Text Generation with Pretrained Language Models"
18 stars 10 forks source link

Pretrained Embedding #5

Open LemonQC opened 3 years ago

LemonQC commented 3 years ago

image

Could you provide the embedding?

turboLJY commented 3 years ago

We utilized the embeddings pretrained by OpenKE (https://github.com/thunlp/OpenKE).

LittleRedLynn commented 3 years ago

Hi, Saker. How did u get the "transe_dict.pkl" and "transe_embedding.npy" files? I download the pretrained embeddings from thunlp website but the files seem not match the ones in the source code of this repo, could you give me some hint? Warmly thanks.

turboLJY commented 3 years ago

You should run TransE in OpenKE based on our datasets to obtain the entity embeddings.

jiangliqin commented 2 years ago

We utilized the embeddings pretrained by OpenKE (https://github.com/thunlp/OpenKE).

i wanna use Chinese corpus to finetune,can I user the OpenKE pretrained embeddings?

zzw-zwzhang commented 2 years ago

same question

zzw-zwzhang commented 2 years ago

Did you solve the problem? Can you provide some detailed guidance? Thanks!

spwidodo commented 1 year ago

Hi

I have runned TransE. But i just got transe.ckpt file.. There is no "transe_dict.pkl" and "transe_embedding.npy" files.. Any solution for this? Thanks ..

ching-0105 commented 1 month ago

Hi

I have runned TransE. But i just got transe.ckpt file.. There is no "transe_dict.pkl" and "transe_embedding.npy" files.. Any solution for this? Thanks ..

Has anyone solved it yet? please...