Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / Unbabel/OpenKiwi issues and pull requests
#116 - What languages do the OpenKiwi support?
Issue -
State: open - Opened by wangpichao 4 months ago
#115 - Why does it need "--model" paramter when I give a specific config?
Issue -
State: open - Opened by michelleqyhqyh almost 2 years ago
- 2 comments
Labels: bug
#114 - What are source_pos and target_pos in the train_config.yaml?
Issue -
State: open - Opened by zolastro almost 2 years ago
#113 - Pretrain config file
Issue -
State: open - Opened by zolastro about 2 years ago
#112 - Update predictor.yaml
Pull Request -
State: open - Opened by zolastro about 2 years ago
- 1 comment
#111 - Do you need to tokenize your data when using a BERT/ROBERTA model?
Issue -
State: open - Opened by zolastro about 2 years ago
#110 - PicklingError: Can't pickle <class 'kiwi.data.encoders.wmt_qe_data_encoder.InputFields[PositiveInt]'>: attribute lookup InputFields[PositiveInt] on kiwi.data.encoders.wmt_qe_data_encoder failed
Issue -
State: open - Opened by Blinkblade over 2 years ago
- 2 comments
Labels: bug
#109 - Seems that maximum token support for a sentence is 512?
Issue -
State: closed - Opened by Duonghailee over 2 years ago
#108 - Got exception when import kiwi
Issue -
State: open - Opened by gaohuan2015 over 2 years ago
Labels: bug
#107 - pkgutil.iter_modules() error: 'PosixPath' object has no attribute 'startswith'
Issue -
State: open - Opened by mtreviso almost 3 years ago
Labels: bug
#106 - some confusions
Issue -
State: open - Opened by leileilin almost 3 years ago
#105 - Fix bug that always downloads tokenizer files.
Pull Request -
State: open - Opened by yym6472 almost 3 years ago
- 1 comment
#104 - Error Pre-Training Predictor: "model -> encoder -> encode_source extra fields not permitted (type=value_error.extra)"
Issue -
State: open - Opened by Qbop981001 almost 3 years ago
- 1 comment
Labels: bug
#103 - Do openKiwi have confident score?
Issue -
State: open - Opened by baoloc97 about 3 years ago
- 1 comment
Labels: question
#102 - OpenKiwi always download the tokenizer files for XLMRoberta even if a local path is configured.
Issue -
State: open - Opened by yym6472 about 3 years ago
- 2 comments
Labels: bug, enhancement
#101 - The prediction process is not complete by Predictor Estimator.
Issue -
State: closed - Opened by Balakiranred about 3 years ago
- 5 comments
Labels: question
#100 - TypeError: cannot unpack non-iterable NoneType object
Issue -
State: closed - Opened by leileilin about 3 years ago
- 1 comment
Labels: bug
#99 - open cannot unpack non-iterable NoneType object
Issue -
State: closed - Opened by genbei about 3 years ago
- 16 comments
Labels: bug
#98 - OSError: Can't load weights for 'xlm-roberta-base'.
Issue -
State: closed - Opened by genbei about 3 years ago
- 16 comments
Labels: bug
#97 - Error at Predictor Training: "Predictor is not a subclass of QESystem"
Issue -
State: open - Opened by tervingo about 3 years ago
- 2 comments
Labels: question
#96 - I suppose that the code comment should be remove.
Issue -
State: closed - Opened by zh25714 over 3 years ago
- 2 comments
Labels: bug
#95 - some problems about data without alignments
Issue -
State: closed - Opened by leileilin over 3 years ago
- 11 comments
Labels: bug
#94 - some confusions
Issue -
State: closed - Opened by leileilin over 3 years ago
#93 - openkiwi version
Issue -
State: closed - Opened by leileilin over 3 years ago
- 2 comments
Labels: bug
#92 - Poor results when training Estimator with parallel data and TER scores
Issue -
State: open - Opened by lluisg over 3 years ago
- 1 comment
Labels: question
#91 - training failed when only predicting source tags
Issue -
State: closed - Opened by iamhere1 over 3 years ago
- 10 comments
Labels: bug
#90 - Add transformer encoder fine-tuning with Adapters
Pull Request -
State: open - Opened by daandouwe over 3 years ago
- 1 comment
#89 - Fix CI by avoiding conflict between torch and pytorch lightning
Pull Request -
State: closed - Opened by daandouwe over 3 years ago
- 1 comment
#88 - Add transformer encoder fine-tuning with Adapters
Issue -
State: open - Opened by daandouwe over 3 years ago
Labels: enhancement
#87 - Add calibration and active learning with Baal
Issue -
State: open - Opened by daandouwe over 3 years ago
Labels: enhancement
#86 - Add config example to docs
Pull Request -
State: closed - Opened by daandouwe over 3 years ago
#85 - Fix doc linking
Pull Request -
State: closed - Opened by captainvera over 3 years ago
Labels: bug
#84 - Relative links break when displayed in docs
Issue -
State: closed - Opened by captainvera over 3 years ago
Labels: bug
#83 - Fix evaluate pipeline by adding verbose and quiet to config
Pull Request -
State: closed - Opened by daandouwe over 3 years ago
Labels: bug
#82 - Evaluate pipeline cannot deal with `verbose` and `quiet` arguments
Issue -
State: closed - Opened by daandouwe over 3 years ago
Labels: bug
#81 - Error happened when running demo
Issue -
State: closed - Opened by iamhere1 over 3 years ago
- 7 comments
Labels: bug
#80 - Add config example
Pull Request -
State: closed - Opened by daandouwe over 3 years ago
Labels: bug
#79 - Unable to start openkiwi
Issue -
State: closed - Opened by knikolaou48 over 3 years ago
- 1 comment
Labels: bug
#78 - Remove extra option that is never used
Pull Request -
State: closed - Opened by captainvera over 3 years ago
Labels: bug
#77 - Add alignment prediction with SimAlign
Issue -
State: open - Opened by daandouwe over 3 years ago
Labels: enhancement
#76 - Fix train from file
Pull Request -
State: closed - Opened by captainvera over 3 years ago
- 2 comments
Labels: bug
#75 - TypeError: metaclass conflict: the metaclass of a derived class must be a (non-strict) subclass of the metaclasses of all its bases
Issue -
State: closed - Opened by Shannen3206 over 3 years ago
- 4 comments
Labels: bug
#74 - Update bug_report.md
Pull Request -
State: closed - Opened by Shannen3206 over 3 years ago
- 2 comments
#73 - Increase test coverage
Pull Request -
State: closed - Opened by daandouwe over 3 years ago
- 1 comment
#72 - Add Streamlit demo
Pull Request -
State: closed - Opened by daandouwe over 3 years ago
#71 - Add Streamlit demo
Issue -
State: closed - Opened by daandouwe over 3 years ago
- 1 comment
Labels: enhancement
#70 - Add search with Optuna
Pull Request -
State: closed - Opened by daandouwe over 3 years ago
- 1 comment
Labels: enhancement
#69 - Pre-trained models for v. 2.0.0
Issue -
State: closed - Opened by andrasaponyi over 3 years ago
#68 - Release Kiwi 2.0.0
Pull Request -
State: closed - Opened by captainvera almost 4 years ago
#67 - Hyperparamete tuning using skorch
Issue -
State: closed - Opened by danparamov almost 4 years ago
- 2 comments
Labels: question
#66 - I have use quickstart .
Issue -
State: closed - Opened by cyjack about 4 years ago
- 2 comments
#65 - Replacing the predictor by BERT or XLM to implement PREDEST-BERT and PREDEST-XLM?
Issue -
State: closed - Opened by AriaChen about 4 years ago
- 7 comments
Labels: enhancement
#64 - Vocabulary size for predictor training
Issue -
State: closed - Opened by andrasaponyi about 4 years ago
#63 - How can I continue my predictor training when interrupted?
Issue -
State: closed - Opened by hcdeng6 about 4 years ago
- 2 comments
#62 - Documentation: Can't find yaml configuration files for QUETCH
Issue -
State: closed - Opened by warlock2k about 4 years ago
- 6 comments
Labels: question
#61 - Add "to" method to move Predicter objects to other devices (#60)
Pull Request -
State: closed - Opened by captainvera over 4 years ago
- 1 comment
#60 - Error in predicting when using GPU for predictor-estimator
Issue -
State: closed - Opened by prabhakar267 over 4 years ago
- 9 comments
Labels: bug
#59 - Unable to predict gap tags
Issue -
State: closed - Opened by zouharvi over 4 years ago
- 5 comments
Labels: bug
#58 - Fix minor documentation error
Pull Request -
State: closed - Opened by pixelneo over 4 years ago
- 4 comments
#57 - Question about data
Issue -
State: closed - Opened by mzeidhassan over 4 years ago
- 4 comments
#56 - Can i use openkiwi to do data cleaning?
Issue -
State: closed - Opened by BigBorg over 4 years ago
- 2 comments
#55 - how fast is the predictor training supposed to be?
Issue -
State: closed - Opened by BigBorg over 4 years ago
- 2 comments
#54 - Hi, I am trying to load the model once and then predict to stdout as the input is fed to stdin. I cannot find any module in the python package to do this.
Issue -
State: closed - Opened by Bhavani01 over 4 years ago
- 4 comments
Labels: enhancement
#53 - Different servers to load the trained PredEst model issue
Issue -
State: closed - Opened by HanchengDeng over 4 years ago
- 4 comments
Labels: bug
#52 - Fix poetry setting that was breaking circleCI checks
Pull Request -
State: closed - Opened by captainvera over 4 years ago
Labels: bug
#51 - Fix formating error with parameter logging
Pull Request -
State: closed - Opened by captainvera over 4 years ago
- 2 comments
Labels: bug
#50 - Add model parameters log to mlflow
Pull Request -
State: closed - Opened by francoishernandez over 4 years ago
- 3 comments
#49 - Add run_name option to make mlflow exps more identifiable
Pull Request -
State: closed - Opened by francoishernandez over 4 years ago
#48 - Fix poetry command at the end of install doc
Pull Request -
State: closed - Opened by francoishernandez over 4 years ago
- 2 comments
#47 - array must not contain infs or NaNs
Issue -
State: closed - Opened by BigBorg over 4 years ago
- 7 comments
Labels: bug
#46 - Is it possible to train with just src, mt, ter?
Issue -
State: closed - Opened by jimbogill over 4 years ago
- 3 comments
Labels: question
#45 - Fix docs of loading predictor when training estimator
Pull Request -
State: closed - Opened by captainvera over 4 years ago
#44 - Fix device assignment for pytorch compatibility solving #43
Pull Request -
State: closed - Opened by captainvera over 4 years ago
#43 - GPU training crashes
Issue -
State: closed - Opened by Bhavani01 over 4 years ago
- 36 comments
Labels: bug
#42 - Plan to release other pre-trained models?
Issue -
State: closed - Opened by sorami almost 5 years ago
- 4 comments
#41 - Update default train estimator config
Pull Request -
State: closed - Opened by zouharvi almost 5 years ago
- 2 comments
#40 - Word level QE reproduction
Issue -
State: closed - Opened by zouharvi almost 5 years ago
- 3 comments
Labels: bug
#39 - Fix compatibility with PyTorch 1.2
Pull Request -
State: closed - Opened by kepler almost 5 years ago
#38 - Incompatibility with Torch 1.2
Issue -
State: closed - Opened by soares-f almost 5 years ago
- 1 comment
Labels: bug
#37 - Fix mlflow logger to support newest version
Pull Request -
State: closed - Opened by captainvera almost 5 years ago
#36 - Add saving of latest model during training and improved error handling of resuming
Pull Request -
State: closed - Opened by captainvera almost 5 years ago
- 1 comment
Labels: enhancement
#35 - Failed to rerun predictor model
Issue -
State: closed - Opened by damyana79 about 5 years ago
- 4 comments
Labels: bug
#34 - `inf` value for RMSE metric
Issue -
State: closed - Opened by edufierro about 5 years ago
- 3 comments
Labels: bug
#33 - Implement Multi-GPU training
Issue -
State: open - Opened by captainvera about 5 years ago
- 5 comments
Labels: enhancement
#32 - predictor-estimator crashes with Russian data
Issue -
State: closed - Opened by ninalopatina about 5 years ago
- 1 comment
Labels: bug
#31 - GPU ID
Issue -
State: closed - Opened by Parkchanjun about 5 years ago
- 2 comments
Labels: bug
#30 - GL integration testing
Issue -
State: closed - Opened by captainvera about 5 years ago
#29 - IndexError: tuple index out of range while using sentence-level Predictor-Estimator to train
Issue -
State: closed - Opened by BUAAers about 5 years ago
- 10 comments
Labels: bug
#28 - Artificially generated triplets?
Issue -
State: closed - Opened by MiroFurtado about 5 years ago
- 6 comments
Labels: question
#27 - Questions about prediction results and evaluate module
Issue -
State: closed - Opened by HXX97 about 5 years ago
- 3 comments
Labels: good first issue
#26 - Where is ape-qe and all its files?
Issue -
State: closed - Opened by HuihuiChyan about 5 years ago
- 3 comments
#25 - Can not find all five models.
Issue -
State: closed - Opened by HuihuiChyan about 5 years ago
- 1 comment
#24 - missing WMT 2017 word_level/test.tags required for predictor-estimator evaluation
Issue -
State: closed - Opened by ninalopatina about 5 years ago
- 4 comments
Labels: bug
#23 - Error occurred while using sentence-level Predictor-Estimator to predict
Issue -
State: closed - Opened by Zachary-YL about 5 years ago
- 5 comments
Labels: bug
#22 - Failed to conduct Predictor-Estimator predicting
Issue -
State: closed - Opened by lihongzheng-nlp about 5 years ago
- 13 comments
Labels: bug
#21 - Add simple visualization Notebook.
Pull Request -
State: closed - Opened by trenous about 5 years ago
#20 - Combining NuQE predictions for words and gaps
Issue -
State: closed - Opened by erickrf over 5 years ago
- 1 comment
Labels: enhancement
#18 - Error training estimator model
Issue -
State: closed - Opened by HqWu-HITCS over 5 years ago
- 18 comments
Labels: bug
#17 - Add logging options to documentation
Pull Request -
State: closed - Opened by captainvera over 5 years ago
Labels: bug
#15 - Output format names are somewhat confusing
Issue -
State: closed - Opened by erickrf over 5 years ago
- 1 comment
Labels: enhancement