OpenNMT/OpenNMT issues and pull requests

#579 - ModuleNotFoundError: No module named 'onmt.inputters'

Issue - State: closed - Opened by Roweida-Mohammed over 4 years ago - 1 comment

#578 - Updating the intel-mkl URL.

Pull Request - State: closed - Opened by arturgontijo over 4 years ago - 2 comments

#575 - NMT, What if we do not pass input for decoder?

Issue - State: closed - Opened by guotong1988 about 5 years ago - 1 comment

#573 - Is `early-stopping` available feature in OpenNMT-tf or OpenNMT-py?

Issue - State: closed - Opened by byuns9334 about 5 years ago

#572 - Is early stopping available in OpenNMT-py or OpenNMT-tf?

Issue - State: closed - Opened by byuns9334 about 5 years ago - 1 comment

#569 - Create CODE_OF_CONDUCT.md

Pull Request - State: open - Opened by Warkanlock over 5 years ago - 1 comment

#568 - Error BPE tokenizer: tokenize.lua:87: attempt to concatenate upvalue 'line'

Issue - State: open - Opened by i55code over 5 years ago - 3 comments

#567 - unicode issue for translating Bangla to English

Issue - State: closed - Opened by AridHasan over 5 years ago - 2 comments

#566 - Maximum length of string processed by the engine

Issue - State: open - Opened by sdlmw over 5 years ago - 2 comments

#565 - Better pre-trained PyTorch models

Issue - State: open - Opened by KeithCu over 5 years ago

#564 - Dockerfile sentencepiece fix

Pull Request - State: closed - Opened by arturgontijo almost 6 years ago - 2 comments

#563 - share_embedding option

Issue - State: closed - Opened by pdc-kaminaga almost 6 years ago - 2 comments

#562 - Test BLEU Score

Issue - State: closed - Opened by MukundKhandelwal almost 6 years ago - 1 comment

#561 - Embeddings: invalid line count

Issue - State: closed - Opened by DiegoMoussallem about 6 years ago - 3 comments

#560 - How to speed up when preprocess the corpus ?

Issue - State: closed - Opened by stevenyslins about 6 years ago - 1 comment

#559 - The engine caches the data in the translation

Issue - State: open - Opened by sdlmw about 6 years ago - 6 comments

#558 - IDEA: Process FOSS projects .po files for translation corpora

Issue - State: open - Opened by Bugsbane about 6 years ago

#557 - WriteObject problem, bad argument #1 to 'ipairs'

Issue - State: closed - Opened by DiegoMoussallem about 6 years ago - 1 comment

#556 - Difference bteween gnmt and brnn encoder

Issue - State: closed - Opened by Anupama94 about 6 years ago - 2 comments

#555 - Difference between using epoch and train_steps

Issue - State: closed - Opened by starnleymbote about 6 years ago - 3 comments

#554 - Meet with lua:446:'for' limit must be a number

Issue - State: closed - Opened by linhy1993 about 6 years ago - 2 comments

#553 - Error when translating with Language Model: "attempt to call method 'forwardOne' (a nil value)"

Issue - State: closed - Opened by indam23 about 6 years ago - 2 comments

#552 - Decoder.lua:446: 'for' limit must be a number

Issue - State: closed - Opened by YshenorTheBis about 6 years ago - 1 comment

#551 - Error when training NMT with language model "for limit must be a number"

Issue - State: closed - Opened by indam23 about 6 years ago - 1 comment

#550 - "

Issue - State: closed - Opened by indam23 about 6 years ago

#549 - Generating pre-trained word embeddings

Issue - State: open - Opened by carolscarton over 6 years ago - 3 comments

#548 - Poor translation result with English->German and German->English pre-trained models

Issue - State: closed - Opened by vardaan123 over 6 years ago - 1 comment

#547 - Segmentation fault in preprocess

Issue - State: closed - Opened by fodilB over 6 years ago - 4 comments

#546 - language detection and translation

Issue - State: open - Opened by lchunleo over 6 years ago

#545 - Fully ignore placeholders when learning BPE

Pull Request - State: closed - Opened by guillaumekln over 6 years ago - 2 comments

#544 - error compiling language model

Issue - State: closed - Opened by jsenellart over 6 years ago - 3 comments

#543 - Add hook for additional mecab tokenization after lua tokenization.

Pull Request - State: open - Opened by Depapepe over 6 years ago - 3 comments

#542 - [WIP] convert to short and introduce padding in the tensor storage

Pull Request - State: open - Opened by jsenellart over 6 years ago - 1 comment

#541 - Illegal memory access after creating SRU modules on CPU and GPU in the same script

Issue - State: closed - Opened by felixgwu over 6 years ago - 1 comment

#540 - Fat Encoder, Thin decoder

Pull Request - State: open - Opened by jsenellart over 6 years ago - 1 comment

#539 - Return source and target tokens with attention table

Pull Request - State: closed - Opened by panosk over 6 years ago - 3 comments

#538 - multi-GPU

Issue - State: open - Opened by wanghechong over 6 years ago - 2 comments

#537 - [Help Wanted] Detailed Explanation of How Word Features Are Predicted

Issue - State: closed - Opened by howardyclo over 6 years ago - 2 comments

#536 - Need help, getting RuntimeError

Issue - State: closed - Opened by yelwinsoe over 6 years ago

#535 - typofix while reading docs

Pull Request - State: closed - Opened by bartvanhalder over 6 years ago - 2 comments

#534 - [Help Wanted] How to apply BPE when input contains word features?

Issue - State: open - Opened by howardyclo over 6 years ago - 9 comments
Labels: enhancement

#533 - Send a string as input instead of file on translate.lua

Issue - State: closed - Opened by grlouk over 6 years ago - 18 comments

#532 - [Multi-task Seq2Seq] Does someone get performance improved when predicting multi-label in target-side?

Issue - State: closed - Opened by howardyclo over 6 years ago - 6 comments

#531 - [Bug] Cannot save validation translation during training

Issue - State: closed - Opened by howardyclo over 6 years ago - 1 comment

#530 - Bug in preprocessing. It creates only two shards. a small and a huge (3GB) one.

Issue - State: closed - Opened by nimasnjb over 6 years ago - 2 comments

#529 - Question: transfer learning

Issue - State: closed - Opened by yapingzhao over 6 years ago - 1 comment

#528 - [WIP] Label smoothing

Pull Request - State: open - Opened by jsenellart over 6 years ago - 1 comment

#527 - Updating ensemble branch to be compatible with current OpenNMT version

Issue - State: closed - Opened by RikVN over 6 years ago - 1 comment

#526 - Problems with setting up lua server in latest docker image

Issue - State: closed - Opened by HuiyingLi over 6 years ago - 2 comments

#525 - Slow at translation time with CPU model

Issue - State: closed - Opened by HuiyingLi over 6 years ago - 3 comments

#524 - Max 1800 tokens per batch?

Issue - State: closed - Opened by getao over 6 years ago - 1 comment

#523 - Does OpenNMT dockerfile provide GPU train when using -gpuid ?

Issue - State: closed - Opened by stevenyslins over 6 years ago - 2 comments

#522 - luarocks install fail when using docker

Issue - State: closed - Opened by stevenyslins over 6 years ago - 1 comment

#521 - Need help to release a pre-trained model from a GPU server

Issue - State: open - Opened by hoangyenan over 6 years ago - 8 comments

#520 - Error in `<path>/torch/install/bin/lua': double free or corruption (!prev): 0x00000000013341b0

Issue - State: open - Opened by laklaja over 6 years ago

#519 - Reduce the Docker image size

Pull Request - State: closed - Opened by guillaumekln over 6 years ago - 1 comment

#518 - The definition of the score outputted by lm.lua score

Issue - State: closed - Opened by getao over 6 years ago - 1 comment

#517 - introduce log_tag option

Pull Request - State: closed - Opened by jsenellart over 6 years ago

#516 - Is there any way to use model ensemble for prediction?

Issue - State: closed - Opened by getao over 6 years ago - 3 comments

#515 - Request For Filipino Translation (fil_PH)

Issue - State: closed - Opened by codygee237 over 6 years ago

#514 - How to deal with huge amounts of training data

Issue - State: closed - Opened by getao over 6 years ago - 4 comments

#513 - Request for translation

Issue - State: closed - Opened by jbeguna04 over 6 years ago

#512 - Release model - error: unknown Torch class <onmt.CustomizedAttention>

Issue - State: open - Opened by jkcchan over 6 years ago - 3 comments

#511 - invalid conversion of ￨n corrected

Pull Request - State: closed - Opened by jsenellart over 6 years ago

#510 - -segment_numbers option

Issue - State: open - Opened by qutie75 over 6 years ago - 1 comment

#509 - fix case feature in data test set

Pull Request - State: closed - Opened by jsenellart over 6 years ago

#508 - How to make the decoder only use the vocab from the input sequence?

Issue - State: open - Opened by wolfshow over 6 years ago - 2 comments

#507 - Can shallow fusion support arbitrary language models other than RNNLM?

Issue - State: open - Opened by getao over 6 years ago - 2 comments

#506 - WIP introduce token batch size limit

Pull Request - State: closed - Opened by vince62s over 6 years ago - 6 comments

#505 - display oov rates

Pull Request - State: closed - Opened by jsenellart over 6 years ago - 1 comment

#504 - Fix492 - complete

Pull Request - State: closed - Opened by jsenellart over 6 years ago

#503 - handle error in opening output files

Pull Request - State: closed - Opened by jsenellart over 6 years ago - 1 comment

#502 - Handle missing target vocabulary for update_vocab

Pull Request - State: closed - Opened by guillaumekln over 6 years ago - 1 comment

#501 - Gbs v2

Pull Request - State: closed - Opened by natsegal over 6 years ago - 1 comment

#500 - Can translate.lua provide an option that allows the model to output n-best sents with pred score?

Issue - State: open - Opened by getao over 6 years ago

#499 - -train_from option doesn't support -model_type lm????

Issue - State: closed - Opened by getao over 6 years ago - 4 comments

#498 - A lot of repetition in the translation results

Issue - State: open - Opened by sdlmw over 6 years ago - 3 comments

#497 - How to feed word features during model inference

Issue - State: closed - Opened by getao over 6 years ago - 2 comments
Labels: question

#496 - Enable evaluation mode for language models

Pull Request - State: closed - Opened by guillaumekln over 6 years ago - 1 comment

#495 - Inference result is not deterministic when decoding with a language model.

Issue - State: closed - Opened by getao over 6 years ago - 1 comment

#494 - data of type xml can be loaded in this project

Issue - State: closed - Opened by sondos90 over 6 years ago - 1 comment

#493 - add option to score on detokenized tests

Pull Request - State: closed - Opened by jsenellart over 6 years ago - 1 comment

#492 - Error at translation

Issue - State: closed - Opened by i55code over 6 years ago - 2 comments

#491 - How to evaluate exsiting model with other metric?

Issue - State: closed - Opened by azureskyL over 6 years ago - 4 comments

#490 - How to improve translation accuracy?

Issue - State: closed - Opened by mewturn over 6 years ago - 2 comments

#485 - Model saving failed after the first epoch

Issue - State: open - Opened by wolfshow over 6 years ago - 5 comments

#483 - Ask for help，I'm having a problem.

Issue - State: open - Opened by sdlmw over 6 years ago - 4 comments

#476 - error with onmt.utils.ExtendedCmdLine.fileExists

Issue - State: closed - Opened by silverguo almost 7 years ago - 2 comments

#473 - pretrained CPU model on opennmt.net is very bad at summarization

Issue - State: closed - Opened by freakeinstein almost 7 years ago - 3 comments

#445 - lm.lua score gives different scores for the same string

Issue - State: closed - Opened by VincentCCL almost 7 years ago

#440 - translation result does not make sense ???

Issue - State: open - Opened by sondos90 almost 7 years ago - 2 comments

#432 - OpenNMT not working on Python in Windows

Issue - State: closed - Opened by LoekvanKooten almost 7 years ago - 1 comment

#400 - Ensemble decoding

Issue - State: closed - Opened by raymondhs almost 7 years ago - 1 comment
Labels: question

#383 - ERROR: all data sources do not have the same number of sentences

Issue - State: closed - Opened by urialon about 7 years ago - 9 comments

#359 - Update options

Pull Request - State: closed - Opened by srush about 7 years ago - 2 comments

#350 - "Not enough memory" error in multi-GPU synchronous parallel training

Issue - State: open - Opened by xiaoda99 about 7 years ago - 2 comments

#300 - [WIP] Ensemble decoding

Pull Request - State: closed - Opened by guillaumekln over 7 years ago - 2 comments

#284 - translation_server.lua:{"error":"JoinTable.lua:38: inconsistent tensor size at THTensorCopy.c:17"}

Issue - State: closed - Opened by incasedo over 7 years ago - 5 comments

#128 - How could I get the score of the attention while translate?

Issue - State: closed - Opened by hfxunlp over 7 years ago - 4 comments

#100 - [WIP] splitting trainer logic

Pull Request - State: closed - Opened by jsenellart over 7 years ago - 5 comments

GitHub / OpenNMT/OpenNMT issues and pull requests