Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / joeynmt/joeynmt issues and pull requests

#228 - the processed data

Issue - State: open - Opened by cecilialeo77 6 months ago - 1 comment

#227 - Implementing Knowledge distillation

Issue - State: open - Opened by vmenan 11 months ago - 3 comments
Labels: question

#226 - enable DDP in inference

Pull Request - State: open - Opened by may- about 1 year ago

#225 - v2.3.0

Pull Request - State: closed - Opened by may- about 1 year ago

#224 - serving & ONNX compat ?

Issue - State: open - Opened by flutter-painter about 1 year ago

#223 - Unit Test Fails - Windows Installation

Issue - State: closed - Opened by jowodo about 1 year ago - 4 comments

#222 - Columns and DataType Not Explicitly Set on line 387 of datasets.py

Issue - State: closed - Opened by CodeSmileBot about 1 year ago
Labels: wontfix

#221 - Update data.py

Pull Request - State: closed - Opened by jowodo over 1 year ago - 1 comment

#220 - Tutorial - Test Set Evaluation

Issue - State: closed - Opened by jowodo over 1 year ago - 5 comments

#219 - Link in Tutorial to Collab dead

Issue - State: closed - Opened by jowodo over 1 year ago - 4 comments
Labels: help wanted

#218 - RFC: Add early stopping for all LR schedulers based on validation score improvements

Pull Request - State: open - Opened by mcognetta over 1 year ago - 1 comment

#217 - Add AdamW Optimizer

Pull Request - State: closed - Opened by mcognetta over 1 year ago - 3 comments

#216 - Ensure correct model tokenizer directory structure exists prior to copying

Pull Request - State: open - Opened by mcognetta over 1 year ago - 5 comments

#215 - Early stopping criteria is only checked for the `ReduceLROnPlateau` scheduler

Issue - State: open - Opened by mcognetta over 1 year ago - 5 comments
Labels: enhancement, work in process

#214 - Basic iwslt config train failure due to directory errors

Issue - State: closed - Opened by mcognetta over 1 year ago - 1 comment

#213 - (enhancement) Deploying trained models on HuggingFace Space

Issue - State: closed - Opened by sinaahmadi over 1 year ago - 2 comments

#211 - make activation fn configurable

Pull Request - State: closed - Opened by may- almost 2 years ago

#210 - RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

Issue - State: closed - Opened by matthew-burnard almost 2 years ago - 5 comments
Labels: duplicate

#209 - Running build_vocab.py for wmt17_bpe with or without --joint?

Issue - State: closed - Opened by AlexWang1900 almost 2 years ago - 3 comments

#208 - trg_mask generate problem

Issue - State: closed - Opened by tongye98 almost 2 years ago - 4 comments
Labels: question

#207 - fix(datasets): retype split

Pull Request - State: closed - Opened by AmitMY almost 2 years ago - 1 comment

#206 - fix unittest #205

Pull Request - State: closed - Opened by may- about 2 years ago

#205 - Unit test FAIL: testSentencepieceTokenizer (test.unit.test_tokenizer.TestTokenizer)

Issue - State: closed - Opened by jeiea about 2 years ago - 4 comments

#204 - fix pylint 2.16.0 errors

Pull Request - State: closed - Opened by may- about 2 years ago

#203 - update quick start tutorial to v2.2

Pull Request - State: closed - Opened by may- about 2 years ago

#202 - JoeyNMT v2.2.0

Pull Request - State: closed - Opened by may- about 2 years ago

#201 - AttributeError: module 'packaging' has no attribute 'version'

Issue - State: closed - Opened by hiro0141mi about 2 years ago - 2 comments

#199 - Move trg_length in batch to device

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

#197 - update joey to v2.1.0 in demo notebook

Pull Request - State: closed - Opened by may- over 2 years ago

#196 - better config validation

Issue - State: open - Opened by may- over 2 years ago
Labels: enhancement, help wanted

#195 - JoeyNMT v1 procedure is no more compatible with JoeyNMT v2

Issue - State: closed - Opened by Nibedita-Roy over 2 years ago - 12 comments
Labels: question, wontfix

#194 - joeynmt v2.1.0

Pull Request - State: closed - Opened by may- over 2 years ago - 1 comment

#193 - Multi-GPU training.

Issue - State: closed - Opened by tongye98 over 2 years ago - 5 comments
Labels: enhancement, help wanted

#192 - Better Transformer `TokenBatchSampler`

Issue - State: open - Opened by AmitMY over 2 years ago - 4 comments
Labels: enhancement, help wanted

#191 - feat(data): allow overriding `collate_fn` without overriding `make_iter`

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

#190 - feat(training): support having no `batch_sampler`

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

#189 - feat(dataset): move data iter creation to dataset class

Pull Request - State: closed - Opened by AmitMY over 2 years ago - 1 comment

#188 - Regression: custom batches no longer supported

Issue - State: closed - Opened by AmitMY over 2 years ago - 17 comments

#187 - install instructions incompatibility with build_vocab.py

Issue - State: closed - Opened by wjbyrne over 2 years ago - 9 comments

#186 - Assertion error on build_vocab.py

Issue - State: closed - Opened by kritisingh24 over 2 years ago - 5 comments

#185 - Get this error

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

#184 - model.eval()

Issue - State: closed - Opened by tongye98 over 2 years ago - 2 comments

#183 - Empty line handling

Issue - State: open - Opened by juliakreutzer over 2 years ago - 5 comments
Labels: bug, work in process

#182 - n_best IndexError

Issue - State: closed - Opened by juliakreutzer over 2 years ago - 2 comments
Labels: bug

#181 - build_vocab.py refactoring

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

#180 - fix for quick start tutorial

Pull Request - State: closed - Opened by may- over 2 years ago

#179 - joeynmt/scripts/build_vocab.py error

Issue - State: closed - Opened by KashyapKishore over 2 years ago - 11 comments

#178 - Newly face this error. Help me

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

#177 - Error

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

#176 - accept datasets.load_from_disk()

Pull Request - State: closed - Opened by juliakreutzer over 2 years ago

#175 - Bpe voc_limit

Pull Request - State: closed - Opened by juliakreutzer over 2 years ago

#174 - JoeyNMT version2.0.0

Pull Request - State: closed - Opened by may- over 2 years ago - 1 comment

#173 - Any scope of Active Learning?

Issue - State: closed - Opened by neerajvashistha over 2 years ago

#172 - Stopping and resuming training

Issue - State: closed - Opened by tarekzaki115 over 2 years ago - 1 comment

#171 - How can I add a new field into the dataset?

Issue - State: closed - Opened by xuda1998 almost 3 years ago

#170 - add files for postnorm

Pull Request - State: closed - Opened by sabrinabraendle almost 3 years ago - 1 comment

#169 - add files for prenorm

Pull Request - State: closed - Opened by sabrinabraendle almost 3 years ago - 2 comments

#168 - Torchtext error

Issue - State: closed - Opened by ShadowVicky almost 3 years ago - 6 comments

#167 - Fix translate mode logic in prediction.py

Pull Request - State: closed - Opened by israaar almost 3 years ago - 1 comment

#166 - Joeynmt and Nvidia-100

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

#165 - validation process takes too long time

Issue - State: closed - Opened by skswldndi about 3 years ago - 2 comments

#164 - joeyHF: adapting joeyNMT for Masakhane Benchmark Models on Spaces

Pull Request - State: closed - Opened by chrisemezue about 3 years ago - 11 comments
Labels: help wanted

#163 - added eps=0.0 parameter to plateau learning rate scheduler.

Pull Request - State: closed - Opened by hagmann-statnlp about 3 years ago - 1 comment

#162 - Feature request: resume training without overwriting model dir

Issue - State: open - Opened by cdleong about 3 years ago - 6 comments
Labels: enhancement

#161 - <unk> hypothesis when using transformer with bpe encoding

Issue - State: closed - Opened by evelynkyl about 3 years ago - 2 comments

#160 - Updated six version requirement

Pull Request - State: closed - Opened by Freshia over 3 years ago

#159 - Sharing pre-trained models through Hugging Face

Issue - State: closed - Opened by osanseviero over 3 years ago - 11 comments
Labels: enhancement, help wanted

#157 - sacrebleu.corpus_bleu() issue

Pull Request - State: closed - Opened by amro-kamal over 3 years ago - 1 comment

#156 - Bug fix

Pull Request - State: closed - Opened by may- over 3 years ago - 1 comment

#155 - corpus_bleu() got an unexpected keyword argument 'sys_stream'

Issue - State: closed - Opened by James-Havinga99 over 3 years ago - 19 comments

#154 - Add acknowledgements to README

Pull Request - State: closed - Opened by sjarmero over 3 years ago

#153 - bug fix #136: ckpt saving logic changed

Pull Request - State: closed - Opened by may- almost 4 years ago - 2 comments

#152 - Windows encodings

Issue - State: closed - Opened by juliakreutzer almost 4 years ago - 2 comments
Labels: bug

#151 - Question about (padding) masking in cross-attention with transformers

Issue - State: closed - Opened by m-decoster almost 4 years ago - 3 comments

#150 - Always use encoding="utf-8" with open

Pull Request - State: closed - Opened by ConstantineLignos almost 4 years ago - 1 comment

#149 - Fixes beam search stop criteria

Pull Request - State: closed - Opened by nicola-decao almost 4 years ago - 8 comments

#148 - Potential bug

Issue - State: closed - Opened by nicola-decao almost 4 years ago - 1 comment

#147 - Added wheel to installation

Pull Request - State: closed - Opened by anebz almost 4 years ago

#146 - unittest for nbest prediction

Pull Request - State: closed - Opened by may- almost 4 years ago - 9 comments

#145 - Bug fixes related to issue #144

Pull Request - State: closed - Opened by sjarmero almost 4 years ago - 11 comments

#144 - possibly a bug in PullRequest #139 (N-best candidates)?

Issue - State: closed - Opened by may- almost 4 years ago - 5 comments
Labels: bug

#143 - File translation results different from interactive outputs

Issue - State: closed - Opened by Evandernyoni almost 4 years ago - 7 comments

#142 - Torchtext, Torch and Numpy version upgrades.

Pull Request - State: closed - Opened by Freshia almost 4 years ago - 6 comments

#141 - Skip tests after training

Pull Request - State: closed - Opened by sjarmero almost 4 years ago - 3 comments

#140 - Updated link to iwslt14 data in preprocessing scripts.

Pull Request - State: closed - Opened by nbrgr almost 4 years ago - 1 comment

#139 - N-best candidates

Pull Request - State: closed - Opened by sjarmero almost 4 years ago - 3 comments

#138 - Update README.md

Pull Request - State: closed - Opened by juliakreutzer almost 4 years ago - 2 comments

#137 - Joey NMT for paper Correct Me If You Can

Issue - State: closed - Opened by AmanSinghal927 about 4 years ago - 2 comments
Labels: question

#136 - expected behavior for `keep_last_ckpts = -1`

Issue - State: closed - Opened by may- about 4 years ago - 9 comments
Labels: bug

#135 - Support for pretrained word embeddings

Pull Request - State: closed - Opened by may- about 4 years ago - 5 comments

#134 - Update README.md

Pull Request - State: closed - Opened by mrshu about 4 years ago - 1 comment

#133 - joeyNMT 1.1/Google Colab - uses device:cpu instead of GPU

Issue - State: closed - Opened by cdleong about 4 years ago - 13 comments

#132 - fix RNN Multi GPU error

Pull Request - State: closed - Opened by may- about 4 years ago - 1 comment

#131 - Fix batch class in test

Pull Request - State: closed - Opened by AmitMY about 4 years ago

#129 - get_iwslt14_bpe.sh broken after link changed

Issue - State: closed - Opened by nbrgr about 4 years ago - 8 comments
Labels: bug

#127 - Implements state_dict and load_state_dict functions for custom schedulers

Pull Request - State: closed - Opened by nbrgr about 4 years ago