https://tejas4patel.r-universe.devRecent package updates in tejas4patelR-universehttps://github.com/tejas4patel.pnghttps://tejas4patel.r-universe.devSat, 17 Oct 2026 04:36:02 GMTalgocrat@gmail.com (Tejas Patel)Fine-tune causal language models with LoRA (Low-Rank Adaptation) from R, and answer questions from your own documents. Documents (PDF, Word, HTML, Markdown, spreadsheets, web pages) are read into chunks, retrieved for each question by combined keyword and embedding search, and answered with cited sources; a model can then write training questions from them for review. Tables of prompts and responses work too. Training runs in a background 'Python' process built on Hugging Face 'transformers' and 'peft', with 4-bit loading (QLoRA) so a 7-billion-parameter model trains on an 8 GB GPU, and a 'shiny' app covers the whole workflow. 'Python' dependencies are declared through 'reticulate' and resolved automatically on first use.https://github.com/r-universe/tejas4patel/actions/runs/39003558517Sat, 17 Oct 2026 04:36:02 GMTdragonfarm0.4.0successhttps://tejas4patel.r-universe.devhttps://github.com/tejas4patel/dragonfarmquickstart.Rmdquickstart.htmlQuickstart: from documents to a model, in R2026-09-14 04:29:082026-10-15 01:52:50reinforcement-learning.Rmdreinforcement-learning.htmlReinforcement learning with rewards you can check2026-09-16 19:47:552026-09-27 18:14:00shiny-app.Rmdshiny-app.htmlThe dragonfarm app2026-09-14 04:29:082026-10-16 14:50:22post-training-loop.Rmdpost-training-loop.htmlThe post-training loop: fine-tune, judge, prefer, repeat2026-09-16 19:47:552026-09-27 18:14:00