SamLynnEvans/Transformer
Transformer seq2seq model, program that can build a language translator from parallel corpus
PythonApache-2.0
Issues
- 0
ys = trg[:, 1:].contiguous().view(-1),why do we have to discard the first seq?
#37 opened by Darleen71 - 2
The version
#36 opened by shark803 - 0
file no found,
#35 opened by 220haoqiaoa - 1
What is argument 'k' in translate.py?
#34 opened by shubhamsrivast4u - 7
- 1
Code bug in Beam.py with line 76-78
#33 opened by afarmer2005 - 4
Error
#21 opened by shibin2018 - 7
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!
#27 opened by anasAloklah - 4
- 0
OSError: [E050] Can't find model 'en'. It doesn't seem to be a shortcut link, a Python package or a valid path to a data directory.
#31 opened by zhaoyewei - 2
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!
#30 opened by trra1988 - 4
- 0
what's the shape of the padding mask?
#28 opened by cswwp - 1
A little error in Positional Encoder
#25 opened by MARD1NO - 0
- 1
- 1
- 2
runtime error
#17 opened by nonva - 0
RuntimeError: The size of tensor a (127) must match the size of tensor b (40) at non-singleton dimension 3
#20 opened by shefali1234 - 1
CUDA assert error or out memory error
#19 opened by MehwishFatimah - 0
How can I feed embeddings from XLM- RoBERTa to transformer seq2seq model?
#18 opened by JohnasSolomon - 0
Does it support to load pretrained model
#16 opened by nonva - 0
ModuleNotFoundError: No module named '_regex'
#15 opened by fabrahman - 1
The train command doesn't seem to work for me
#13 opened by bundle-adjuster - 5
error: file not found
#9 opened by Scum1254 - 1
Adding a new layer to this model
#12 opened by liperrino - 0
create_valset argument in train.py
#11 opened by srajan-jha - 1
- 5
- 3
runtime error
#8 opened by xiaohongniua - 0
Run Time Error and Transfer Learning?
#6 opened by ks0m1c - 0
Misinterpreted multi head attention
#4 opened by zolikacsepel - 3
RuntimeError: Expected object of backend CPU but got backend CUDA for argument #2 'other'
#2 opened by fabrahman - 1
Process.py missing?
#1 opened by TheodoreGalanos