How to get embeddings？

[CallMeNumber1]

I wonder what should i do to get poincare-embeddings after ruuning ./train-nouns.sh, which will result in a '.pth' file, and then how to get embeddings from the '.pth' file?

[Matt Le]

This should do it:

import torch
chkpnt = torch.load('<path to checkpoint>')
embeddings = chkpnt['model']['lt.weight']