sillsdev / sillsdev/silnlp

Best selection of verses (golden path) to jump start fine tuning

Open
#183 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

needs to be revisited research
Dominant language
Python
Stars
39
Forks
7
Avg merge
1d 9h
Merged PRs (30d)
5

Description

How to figure out the best selection of verses to help jump start the NLLB model for best translations?

  • Baseline:
    • Use Matthew to train
    • Bleu score off of Mark, John, Genesis and Jonah
  • Algorithm 1: Random selection of verses
    • Make all Bible verses into vectors
    • Start with a selection of "really nice" verses that are helpful for people groups
    • Extend to a full set equal to the number of verses in Matthew (though vector dissimilarity? What other mechanisms?)
    • Train and get Bleu score from the same books
  • Algorithm 2: Rapid word collection with XRI's sentence generation:
    • What does this entail? How much and what type of corpora is needed?

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or entry points are named. Start by defining the Matthew training baseline and the Mark, John, Genesis, and Jonah BLEU evaluation, then compare the proposed verse-selection and XRI sentence-generation approaches. Done means documenting a reproducible selection method and measured comparison.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.