Telegram Group Search
Coolify: Open-source and self-hostable Heroku / Netlify / Vercel alternative (❄️ Score: 151+ in 2 days)

Link: https://readhacker.news/s/6s33E
Comments: https://readhacker.news/c/6s33E
Learn electricity and electronics fundamentals without taking a formal course (🔥 Score: 151+ in 3 hours)

Link: https://readhacker.news/s/6sdNS
Comments: https://readhacker.news/c/6sdNS
Annotated Unix Magic Poster (Score: 152+ in 16 hours)

Link: https://readhacker.news/s/6sdzL
Comments: https://readhacker.news/c/6sdzL
Emulating an iPhone in QEMU (Score: 151+ in 9 hours)

Link: https://readhacker.news/s/6seDT
Comments: https://readhacker.news/c/6seDT
Show HN: OCR pipeline for ML training (tables, diagrams, math, multilingual) (Score: 150+ in 19 hours)

Link: https://readhacker.news/s/6secG
Comments: https://readhacker.news/c/6secG

Hi HN,
I’ve been working on an OCR pipeline specifically optimized for machine learning dataset preparation. It’s designed to process complex academic materials — including math formulas, tables, figures, and multilingual text — and output clean, structured formats like JSON and Markdown.
Some features:
• Multi-stage OCR combining DocLayout-YOLO, Google Vision, MathPix, and Gemini Pro Vision
• Extracts and understands diagrams, tables, LaTeX-style math, and multilingual text (Japanese/Korean/English)
• Highly tuned for ML training pipelines, including dataset generation and preprocessing for RAG or fine-tuning tasks
Sample outputs and real exam-based examples are included (EJU Biology, UTokyo Math, etc.)
Would love to hear any feedback or ideas for improvement.
GitHub: https://github.com/ses4255/Versatile-OCR-Program
2025/04/06 01:25:03
Back to Top
HTML Embed Code: