Issues
- 0
- 3
only spaces get predicted
#18 opened by aadityapattabhiraman - 0
W_o matrix in multiheadattention
#23 opened by PING-An32 - 1
Clarification regarding decoder_input and label
#22 opened by BrLlan - 0
- 0
Not an issue, but a re-implementation
#20 opened by chettiargautam - 2
translate.py not consistent
#19 opened by dzjxzyd - 1
- 0
Getting Error : module transformers has no attribute 'PreTrainedtokenizerBase'
#16 opened by shuklaji28 - 1
Quality after 20 epoch training
#15 opened by thanhnew2001 - 2
- 0
Fourier Positional Embeddings
#13 opened by mourad-ghafiri - 1
Truncation for Tokenizer
#12 opened by ardaaras99 - 2
Position Encoding
#2 opened by Sheiphan - 1
forward method in transformer class
#10 opened by akkasi - 3
dataset
#9 opened by debby-2020 - 3
- 4
Question: train.py
#5 opened by mohamedelbahnasawi - 1
multihead attention biases
#4 opened by stanchiang