pytorch translate issues

pytorch / translate

Translate - a PyTorch Language Library

BSD 3-Clause "New" or "Revised" License

821 stars 191 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Create code.c++

#706 Ayansam1152 opened 3 years ago
1
Replacing Hardswish to alternative activation function

#705 Adhithya-tech opened 3 years ago
0
[*.py] Rename "Arguments:" to "Args:"

#704 SamuelMarks opened 3 years ago
0
Fix for versions of sacrebleu > 1.4.5

#703 mgaido91 closed 3 years ago
1
Allow specifying tokenizer, num samples and sample size when computing BLEU significance

#702 mgaido91 closed 3 years ago
3
Remove pytorch translate tests

#701 cndn closed 3 years ago
2
Add null checks for more stats

#700 cndn closed 4 years ago
2
Fix PPL reporting

#699 cndn closed 4 years ago
2
Remove broken unit test test_beam_search_and_decode_generate

#698 microLizzy closed 4 years ago
2
`build_generator` api changes for the scripted SequenceGenerator (#1922)

#697 liuchen9494 closed 4 years ago
1
Cannot load pretrained model

#696 erip closed 4 years ago
1
Make TransformerDecoupled model scriptable

#695 anchit closed 4 years ago
2
obtain alignment/attention information

#694 lhk closed 4 years ago
1
Fix LevT training on master

#693 cndn closed 4 years ago
1
depricated functions

#692 alrojo closed 4 years ago
4
Pass is_npz to pytorch translate generate

#691 cndn closed 4 years ago
2
Fix NMT test precision caused by mkldnn upgrade

#690 dzhulgakov closed 4 years ago
2
Update CODE_OF_CONDUCT.md

#689 jspisak closed 4 years ago
1
support fairseq (mmap) corpora

#688 jhcross closed 4 years ago
2
test transformer with aan=True param

#687 jhcross closed 4 years ago
2
Bug fix after D19679308

#686 cndn closed 4 years ago
2
cmd-line support for loading mmap datasets

#685 jhcross opened 4 years ago
2
Add validation loss to metric reportor and Remove intermediate eval

#684 cndn opened 4 years ago
1
Script SinusoidalPositionalEmbedding (#1612)

#683 cndn closed 4 years ago
2
Fix generate.py in pytorch_translate for OSS

#682 cndn closed 4 years ago
2
Script MultiheadAttention (#1524)

#681 cndn closed 4 years ago
4
beam search

#680 Oktai15 opened 4 years ago
2
Export for the vocab prediction module

#679 rasoolims closed 3 years ago
2
Learned vocab prediction

#678 rasoolims closed 3 years ago
2
fixes LevenshteinTransformerModel import

#677 erip closed 4 years ago
1
NAT bug fix

#676 cndn opened 4 years ago
1
Kill ONNX export / ForcedDecoder

#675 cndn closed 4 years ago
2
unrecognized arguments: --batched-beam

#674 erip closed 3 years ago
0
NameError: name 'LevenshteinTransformerModel' is not defined

#673 erip closed 4 years ago
3
NAT bug fix

#672 cndn closed 4 years ago
3
Remove fbtranslate C++ code (OS/FB)

#671 cndn closed 4 years ago
3
test scriptify MultiheadAttention

#670 jhcross opened 4 years ago
3
Manifold migration for export code / export code cleanup

#669 cndn closed 4 years ago
6
get attn out from dictionary

#668 qingerVT closed 3 years ago
3
Fix export code after model "hard fork"

#667 cndn closed 4 years ago
3
Point training pipeline to internal version of NAT

#666 cndn closed 4 years ago
3
PathManager for binarized data

#665 jhcross closed 4 years ago
4
Adding character information to the final layer with vocab reduction

#664 rasoolims closed 3 years ago
2
remove export support for transformer_aan arch

#663 jhcross closed 4 years ago
5
Remove batched beam related tests

#662 cndn closed 4 years ago
4
Default quantization for NAT

#661 cndn closed 4 years ago
3
AAN as ptt_transformer option

#660 jhcross closed 4 years ago
3
Flow side support for separate decoders / Fix max_iter at inference

#659 cndn opened 4 years ago
1
Adding character information to the final layer

#658 rasoolims closed 3 years ago
2
Convert matmuls to quantizable nn.Linear modules (#1065)

#657 halilakin closed 4 years ago
1