Skip to content

fix(translations): replace residual Hindi words in Tamil lessons - #744

Open
Som Samantray (SomSamantray) wants to merge 1 commit into
microsoft:mainfrom
SomSamantray:fix-ta-translation-hindi-residue
Open

Som Samantray (SomSamantray) wants to merge 1 commit into
microsoft:mainfrom
SomSamantray:fix-ta-translation-hindi-residue

Conversation

@SomSamantray

Copy link
Copy Markdown

Summary

Machine-translation passes left Hindi (Devanagari) words embedded inside otherwise-Tamil sentences in four files under translations/ta/. Tamil readers hit unreadable Hindi mid-sentence. Each Hindi span is now replaced with its Tamil equivalent, sourced against the English original sentence.

The Tamil README is restored separately by #729; this PR covers the remaining lesson files.

Related: #734

Changes

  • translations/ta/lessons/0-course-setup/how-to-run.mdविभिन्नபல்வேறு ("different Python virtual environments")
  • translations/ta/lessons/4-ComputerVision/08-TransferLearning/TrainingTricks.mdनिर्भरசார்ந்துள்ளது ("depends on the learning rate parameter")
  • translations/ta/lessons/2-Symbolic/README.mdकरणகரண (repairs the Tamil word பிரகரணத்தில்)
  • translations/ta/lessons/6-Other/21-GeneticAlgorithms/Genetic.ipynbनिर्भर करेगाசார்ந்திருக்கும் ("would depend on the fitness function")

Validation

  • Unicode scan: zero Devanagari characters (U+0900–U+097F) remain in the four changed files.
  • Genetic.ipynb passes nbformat.validate; the change touches one markdown cell string only.
  • Each file's diff is a single line; only the Devanagari span changed. No code, links, or English content modified.

AI assistance disclosure

This fix was identified and prepared with AI assistance. All substitutions were verified programmatically against the English source text (Unicode script scan and notebook format validation), per this repository's translation validation guidance.

Machine-translation passes left Devanagari (Hindi) words embedded inside
otherwise-Tamil sentences in four files under translations/ta/. Tamil
readers hit unreadable Hindi mid-sentence (issue microsoft#734). Each substitution
was sourced against the English original:

- 0-course-setup/how-to-run.md: 'विभिन्न' -> 'பல்வேறு' ("different")
- 4-ComputerVision/08-TransferLearning/TrainingTricks.md:
  'निर्भर' -> 'சார்ந்துள்ளது' ("depends")
- 2-Symbolic/README.md: 'करण' -> 'கரண' (repairs the word 'பிரகரணத்தில்')
- 6-Other/21-GeneticAlgorithms/Genetic.ipynb:
  'निर्भर करेगा' -> 'சார்ந்திருக்கும்' ("would depend")

Validation: Unicode scan shows zero Devanagari characters remaining in the
four files; Genetic.ipynb passes nbformat.validate; each file diff is a
single line with only the Devanagari span substituted.

Addresses microsoft#734 (the ta README portion is handled separately by microsoft#729).
@leestott

Copy link
Copy Markdown
Collaborator

Hi Som Samantray (@SomSamantray) translation are auto completed by https://github.com/Azure/co-op-translator please raise translation quality issues on the coop repo

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟢 Approval recommended

The changes are narrowly scoped to translation text fixes, and the updated files contain no remaining Devanagari characters in the modified content.

Pull request overview

This PR cleans up the Tamil (ta) lesson translations by replacing leftover Hindi (Devanagari) words embedded mid-sentence, improving readability for Tamil readers and addressing the issue reported in #734.

Changes:

  • Replaced residual Devanagari spans with appropriate Tamil equivalents across three Markdown lesson files.
  • Updated one Markdown cell string in a Tamil Jupyter notebook translation to remove Devanagari text.
  • Verified (per scan) that no Devanagari characters remain in the four changed files.
File summaries
File Description
translations/ta/lessons/0-course-setup/how-to-run.md Replaces a leftover Hindi word with Tamil in the miniconda/virtual env description.
translations/ta/lessons/4-ComputerVision/08-TransferLearning/TrainingTricks.md Replaces a leftover Hindi word with Tamil in the learning-rate explanation.
translations/ta/lessons/2-Symbolic/README.md Repairs a mixed-script Tamil word by replacing residual Hindi characters with Tamil.
translations/ta/lessons/6-Other/21-GeneticAlgorithms/Genetic.ipynb Updates a markdown cell string to replace a leftover Hindi phrase with Tamil.
Review details
  • Files reviewed: 4/4 changed files
  • Comments generated: 0
  • Review effort level: Lite

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants