mujahid609/Implementation-and-Evaluation-of-a-Basic-Retrieval--Augmented-Generation-RAG-System-in-Medical-field

A concise summary: This repository is a production-ready Medical Retrieval-Augmented Generation (RAG) package for building and testing evidence-backed medical QA systems. It targets ML engineers, researchers, clinicians prototyping NLP tools, and educators who want a runnable end-to-end demo. Key features: - Ready-to-run Colab notebook (Medical_RAG_System_Fixed.ipynb) with valid JSON, 9 labeled cells (install, imports, core classes, docs, model init, query pipeline, Gradio UI, launch, test). - Beautiful, responsive Gradio UI that returns answer + confidence + source docs and supports a documents slider. - Hybrid retrieval: BM25 + semantic embeddings; generation via a RoBERTa-based QA model. - Documentation: USAGE_GUIDE.md, QUICK_REFERENCE.md, SUMMARY.md with examples, troubleshooting, and customization tips. - Developer utility notebook_generator.py to create valid notebooks and avoid JSON parse errors. - Includes 8 sample medical documents, performance specs (first-run model download ~15–20 min, subsequent runs faster), and easy customization (add docs, change models, integrate vector DB).

AI-related keywords found in this repo's description, topics, or README summary — grouped by category. Each badge links to the corresponding ranking detail page.