gordicaleksa/pytorch-original-transformer

Issues

• A environment problem

  #10 opened 9 months ago by BruceWang01
  0

• torchtext.data import not working in the latest versions of pytorch.

  #9 opened a year ago by shreyashkar-ml
  0

• Sorry, but I couldn't understand where is the concatenation layer after the multi head self attention, shouldn't there be?

  #8 opened a year ago by Domics10
  0

• Error when running "python training_script.py --batch_size 100 --dataset_name IWSLT --language_direction G2E

  #6 opened 3 years ago by minertom
  2

• sharing weight matrix between the two embedding layers and the pre-softmax linear transformation

  #7 opened 3 years ago by nataly-obr
  0

• issue when command :python training_script.py --batchsize 2 -- dataset_name IWSLT --language_direction G2E

  #4 opened 4 years ago by adamas-v
  2

• Frequency in the positional encodings

  #5 opened 3 years ago by FAhtisham
  0

• Issue regarding "9.1 Download pretrained transformers automatically"

  #3 opened 4 years ago by CaterinaFabbri
  2

• can you show the bleu for this repo on WMT14 dataset?

  #2 opened 4 years ago by GuangyanZhang
  0