|
If there is any way to keep placeholders as same as source when call NMT
|
|
1
|
3549
|
June 6, 2017
|
|
Paragraph vs sentence segmentation
|
|
2
|
2892
|
October 16, 2017
|
|
Running OpenNMT on Windows - AttributeError: ‘Namespace’ object has no attribute ‘fix_word_vecs_enc’
|
|
1
|
3539
|
September 3, 2021
|
|
-fix_word_vecs_enc and -fix_word_vecs_dec not doing what I expect them to do
|
|
7
|
1768
|
September 16, 2019
|
|
How should I clean the texts before training?
|
|
4
|
2226
|
February 11, 2020
|
|
"-add_tokenized_output "option in rest_translation_server.lua
|
|
8
|
1663
|
May 10, 2018
|
|
Server always returning empty list
|
|
6
|
1882
|
November 16, 2017
|
|
Validating trained model
|
|
6
|
1879
|
May 16, 2017
|
|
Problems with learning rate decay
|
|
4
|
2219
|
June 25, 2019
|
|
Difference between bpe and character-level tokenization for BLUE score
|
|
2
|
2863
|
December 15, 2019
|
|
Unable to train OpenNMT-py on Apple M1
|
|
0
|
880
|
August 11, 2022
|
|
Best model overwrtiten - Shall I use the latest model?
|
|
6
|
1856
|
June 4, 2024
|
|
SentencePiece + pyonmttok
|
|
5
|
1997
|
July 4, 2021
|
|
Text Summarization Question?
|
|
3
|
2443
|
May 2, 2017
|
|
Linker error when compiling CTranslate2 with "-march=corei7-avx" (CXX_FLAGS)
|
|
6
|
1845
|
September 12, 2020
|
|
Data Preparation toy_en_de.yaml
|
|
2
|
2814
|
January 11, 2021
|
|
GPT-2 for french
|
|
4
|
2174
|
January 12, 2021
|
|
GitHub - dothq/translate: 🌐 Translation microservice powered by OpenNMT
|
|
0
|
866
|
November 30, 2021
|
|
Transformer model training not starting
|
|
5
|
1984
|
July 9, 2020
|
|
Tracking Training progress
|
|
1
|
1087
|
December 20, 2021
|
|
‘charmap’ codec can’t encode character ‘\ufffd’ in position 0: character maps to <undefined>
|
|
2
|
2805
|
November 21, 2020
|
|
Translation doesn't continue after specific number of statements
|
|
6
|
1836
|
November 8, 2019
|
|
Problem when training with fp16
|
|
5
|
1983
|
December 9, 2019
|
|
Command errored out with exit status 128: git clone
|
|
0
|
4857
|
January 20, 2020
|
|
Number of samples in one training step together with accumulation
|
|
6
|
1834
|
April 8, 2021
|
|
How to get a dictionary from .pt file?
|
|
1
|
3422
|
November 25, 2019
|
|
Docker Instructions
|
|
3
|
2419
|
April 5, 2017
|
|
Exposure bias during training
|
|
3
|
2417
|
August 10, 2017
|
|
Using SharedEmbeddings Transformer model with Pretrained Embeddings
|
|
0
|
860
|
February 8, 2024
|
|
Assert model_dim % head_count == 0
|
|
1
|
1075
|
November 8, 2022
|
|
Get VMap from the corpus to be used in CTranslate2
|
|
8
|
1602
|
March 24, 2020
|
|
Training crashes at evaluation due to UnicodeDecodeError
|
|
5
|
1954
|
March 30, 2018
|
|
Inferencing with translate.py
|
|
6
|
1811
|
March 9, 2022
|
|
Error with latest CTranslate2 version in OpenNMT-py REST server
|
|
3
|
2392
|
July 5, 2021
|
|
How word features work?
|
|
6
|
1806
|
June 30, 2019
|
|
Cuda OOM for Validation
|
|
8
|
1589
|
December 17, 2018
|
|
How to use Bert embedding into OpenNMT seq2seq model
|
|
4
|
2131
|
May 11, 2021
|
|
Running OpenNMT-tf Quickstart Very Slow
|
|
8
|
1588
|
March 25, 2020
|
|
Multimodal feature input for seq2seq
|
|
7
|
1682
|
August 11, 2021
|
|
Op_kernel.cc:1192 Out of range: Attempted to repeat an empty dataset infinitely
|
|
3
|
2377
|
February 5, 2018
|
|
Visualizing the graph
|
|
7
|
1677
|
June 16, 2020
|
|
Pre-train encoder side UTF-16 word embeddings in OpenNMT-py?
|
|
0
|
844
|
August 2, 2018
|
|
Translate_batch(): incompatible function arguments. for ctranslate2
|
|
1
|
3349
|
January 18, 2023
|
|
Can I use PyOpenNMT to load/translate a model built with Torch OpenNMT?
|
|
4
|
2116
|
June 16, 2017
|
|
Difference in bleu & perplexity scores between almost same trainings
|
|
4
|
2113
|
December 17, 2020
|
|
Using a learned BPE Model for Transformer
|
|
0
|
841
|
March 9, 2024
|
|
EOFError: Source and reference streams have different lengths
|
|
3
|
2355
|
February 22, 2020
|
|
Translate so slow
|
|
4
|
2102
|
July 6, 2021
|
|
Integrating ctranslate2 with Unreal Engine
|
|
0
|
836
|
September 23, 2024
|
|
How to train from sequences of features instead of sentences
|
|
7
|
1661
|
March 27, 2017
|