sillsdev/silnlp

View on GitHub

A set of pipelines for performing experiments on various NLP tasks with a focus on resource-poor/minority languages.

Stars: 39
Forks: 7
Open beginner issues: 0
Indexed issues: 136
Avg merge: 1d 9h
Merged PRs (30d): 5

Dominant language: Python
License: No license data
Last GitHub push: Sep 3, 2026
Latest indexed: Sep 19, 2026
Contributing guide: No contributing guide
Code of conduct: No code of conduct
Beginner labels: good first issue

Beginner friendly Everything

Open All

136 open issues indexed

Alignment score-based filtering is not supported for corpora (only scripture) data files Open

enhancement pipeline 3: preprocess

sillsdev/silnlp#1156 · 1 assignee · Opened Sep 18, 2026
Extra versification error indicating many missing verses Open

bug onboarding

sillsdev/silnlp#1155 · 2 assignees · Opened Sep 18, 2026
During onboarding, the tokenization stats can fail if there are no parallel source/target verses Open

enhancement onboarding

sillsdev/silnlp#1152 · 1 assignee · Opened Sep 17, 2026
Add KBT support for "All" KBT lists Open

enhancement machine pipeline 3: preprocess

sillsdev/silnlp#1141 · 2 comments · 2 assignees · Opened Aug 19, 2026
Explicitly pass versification to machine.py APIs Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

sillsdev/silnlp#1126 · Opened Jul 30, 2026
Add the Machine.py Eflomal aligner as an option for SILNLP's alignment scripts Open

enhancement

Difficulty 3/5 1-2 days Newbie friendliness 68/100

sillsdev/silnlp#1108 · Opened Jul 2, 2026
Marker placement does not work for scripts that do not use spaces Open

Difficulty 3/5 1-2 days Newbie friendliness 62/100

sillsdev/silnlp#1107 · 2 comments · Opened Jul 2, 2026
Only training alignment model for marker placement on translated content Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

sillsdev/silnlp#1102 · Opened Jun 25, 2026
Add GitHub Action to run style checking on core files Open

enhancement mlops

Difficulty 3/5 1-2 days Newbie friendliness 55/100

sillsdev/silnlp#1100 · 1 comment · Opened Jun 19, 2026
Upgrade poetry to latest version Open

mlops

Difficulty 2/5 1-3 hours Newbie friendliness 65/100

sillsdev/silnlp#1099 · Opened Jun 19, 2026
Limitation of using Original versification as standard Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

sillsdev/silnlp#1092 · 1 comment · Opened Jun 8, 2026
Number of total possible verses is versification-specific Open

Difficulty 3/5 1-2 days Newbie friendliness 58/100

sillsdev/silnlp#1091 · 4 comments · Opened Jun 8, 2026
Support mixed-source drafting when one source shares a language code with the target Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 45/100

sillsdev/silnlp#1088 · Opened Jun 4, 2026
onboard_project fails during the extract step if the project's Biblical Terms refer to another project that isn't available on the shared drive Open

bug onboarding pipeline 2: extract

Difficulty 3/5 1-2 days Newbie friendliness 62/100

sillsdev/silnlp#1082 · 1 comment · Opened May 29, 2026
Generated drafts follow the source project's versification; they should follow the target project's versification. Open

bug machine pipeline 6: infer

sillsdev/silnlp#1069 · 1 comment · 2 assignees · Opened May 14, 2026
Allow new tokens to be initialized to an existing token's embeddings Open

enhancement pipeline 4: train

Difficulty 4/5 3-5 days Newbie friendliness 52/100

sillsdev/silnlp#1061 · Opened May 7, 2026
When characters are added to the tokenizer with the init_unk: true setting, the first 2 characters are not initialized with the <unk> embeddings Open

bug pipeline 4: train

sillsdev/silnlp#1060 · 2 assignees · Opened May 6, 2026
Switch `SilNlpEnv` to be injected instead of global Open

refactoring

sillsdev/silnlp#1054 · 1 assignee · Opened May 5, 2026
Upgrade SILNLP to a newer version of Python Open

machine maintenance

sillsdev/silnlp#1046 · 3 comments · 1 assignee · Opened May 1, 2026
Add support for draft quality estimates for configurations with multiple source languages Open

enhancement pipeline 6: infer

sillsdev/silnlp#1045 · 2 comments · 1 assignee · Opened Apr 30, 2026
Consolidate postprocessing scripts Open

pipeline 7: postprocess

Difficulty 4/5 3-5 days Newbie friendliness 48/100

sillsdev/silnlp#1039 · Opened Apr 30, 2026
Remove unused functionality from core parts of SILNLP Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

sillsdev/silnlp#1030 · 1 comment · Opened Apr 30, 2026
Expand coverage of `nmt/experiment.py` smoke test Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 48/100

sillsdev/silnlp#1029 · Opened Apr 30, 2026
Create a smoke test for `onboard_project.py` Open

Difficulty 3/5 1-2 days Newbie friendliness 65/100

sillsdev/silnlp#1028 · Opened Apr 30, 2026
Create a smoke test for `extract_corpora.py` Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

sillsdev/silnlp#1027 · Opened Apr 30, 2026
Create a smoke test for `analyze.py` Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

sillsdev/silnlp#1026 · Opened Apr 30, 2026
Create a smoke test for `nmt/translate.py` Open

Difficulty 2/5 1-3 hours Newbie friendliness 68/100

sillsdev/silnlp#1025 · Opened Apr 30, 2026
Create a smoke test for `nmt/test.py` Open

Difficulty 3/5 1-2 days Newbie friendliness 62/100

sillsdev/silnlp#1024 · Opened Apr 30, 2026
Create a smoke test for `preprocess.py` Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

sillsdev/silnlp#1023 · Opened Apr 30, 2026
Translate step fails if the project's Settings.xml has an empty <TranslationInfo> field Open

bug pipeline 6: infer

sillsdev/silnlp#1019 · 2 comments · 1 assignee · Opened Apr 29, 2026
Torch `autocast` deprecation warning Open

bug

sillsdev/silnlp#1009 · 1 comment · 1 reaction · 2 assignees · Opened Apr 24, 2026
With paragraph break applied, many second lines in Proverbs begin with punctuation marks Open

enhancement pipeline 7: postprocess

Difficulty 4/5 3-5 days Newbie friendliness 42/100

sillsdev/silnlp#996 · Opened Apr 22, 2026
Improve error message when custom versification does not match detected versification Open

Difficulty 3/5 1-2 days Newbie friendliness 55/100

sillsdev/silnlp#988 · Opened Apr 7, 2026
Improve error reporting for USFM versification issues to open file with Ctrl+Click Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

sillsdev/silnlp#982 · Opened Mar 30, 2026
Sequence score includes scores for lang id token Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

sillsdev/silnlp#966 · 6 comments · Opened Mar 13, 2026
Onboard_projects.py places project files outside of new folder in original location only Open

onboarding

sillsdev/silnlp#945 · 1 assignee · Opened Feb 19, 2026
Drafting from a source book with verse split(s) requires updates to the target project's custom.vrs Open

enhancement pipeline 6: infer

Difficulty 5/5 Over a week Newbie friendliness 35/100

sillsdev/silnlp#922 · Opened Jan 28, 2026
Verse segmentation - when confidence scores are low, combine two verses into a verse range. Open

enhancement pipeline 3: preprocess research

Difficulty 5/5 Over a week Newbie friendliness 25/100

sillsdev/silnlp#916 · Opened Jan 24, 2026
Retrospective research opportunities and forward-thinking testing/engagement using ClearML data Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

sillsdev/silnlp#915 · Opened Jan 23, 2026
Confidence feature not working on test.py for --multiple-translations Open

bug pipeline 5: test

Difficulty 3/5 1-2 days Newbie friendliness 45/100

sillsdev/silnlp#911 · 2 comments · Opened Jan 17, 2026
Key Terms end-user documentation Open

documentation

sillsdev/silnlp#902 · 2 assignees · Opened Dec 18, 2025
Handle different versifications in segment_verses.py Open

enhancement pipeline 3: preprocess research

Difficulty 5/5 Over a week Newbie friendliness 30/100

sillsdev/silnlp#895 · Opened Dec 9, 2025
No-break spaces (u+00A0) are not used in drafts Open

enhancement pipeline 3: preprocess pipeline 6: infer

Difficulty 4/5 3-5 days Newbie friendliness 35/100

sillsdev/silnlp#873 · 2 comments · Opened Nov 21, 2025
Create drafts that use the project's selected normalization. Open

enhancement onboarding pipeline 6: infer

Difficulty 4/5 3-5 days Newbie friendliness 38/100

sillsdev/silnlp#868 · Opened Nov 17, 2025
Create draft files using the target project's encoding Open

bug pipeline 6: infer

Difficulty 3/5 1-2 days Newbie friendliness 45/100

sillsdev/silnlp#859 · 1 comment · Opened Nov 4, 2025
Get verse text from target and reference for usability sample Open

Difficulty 2/5 1-3 hours Newbie friendliness 58/100

sillsdev/silnlp#858 · Opened Nov 3, 2025
Standardize environment for onboarding users Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

sillsdev/silnlp#838 · Opened Oct 17, 2025
Improved mixing of back translation in drafting configuration Open

enhancement pipeline 3: preprocess

Difficulty 5/5 Over a week Newbie friendliness 25/100

sillsdev/silnlp#837 · Opened Oct 15, 2025
Improvements to extract_flex script Open

pipeline 2: extract

Difficulty 4/5 3-5 days Newbie friendliness 35/100

sillsdev/silnlp#822 · 1 comment · Opened Sep 29, 2025
Allow the onboarding script to perform language identification Open

enhancement onboarding pipeline 2: extract

Difficulty 4/5 3-5 days Newbie friendliness 45/100

sillsdev/silnlp#782 · Opened Jul 31, 2025

sillsdev/silnlp

Get new issues in your inbox