diff --git a/.env.example b/.env.example index d6e79b15de84350ea55a005cbf4098697f32ba07..a484d4a9c7c90d9bcb544c3994d14d09853e4844 100644 --- a/.env.example +++ b/.env.example @@ -1,15 +1,11 @@ -GROQ_API_KEY= AWS_ACCESS_KEY_ID= -AWS_SECRET_ACCESS_KEY= AWS_DEFAULT_REGION= +AWS_SECRET_ACCESS_KEY= +GROQ_API_KEY= HUGGINGFACEHUB_ACCESS_TOKEN= -PORT= - - +PORT=8000 TAVILY_API_KEY= -Gemini_API_Key= HF_TOKEN= -APP_API_KEY= diff --git a/.gitattributes b/.gitattributes index 0f1617ef87851f455307894fff64cb94b8262753..c2441dc3c1ef54982f5feaa4f3c31550d2642716 100644 --- a/.gitattributes +++ b/.gitattributes @@ -46,3 +46,8 @@ images/ml_model_example.png filter=lfs diff=lfs merge=lfs -text images/ml_workflow_diagram.png filter=lfs diff=lfs merge=lfs -text images/transformer_application.png filter=lfs diff=lfs merge=lfs -text images/transformer_architecture.png filter=lfs diff=lfs merge=lfs -text +data/google.docx filter=lfs diff=lfs merge=lfs -text +references/Screenshot[[:space:]]from[[:space:]]2026-05-19[[:space:]]09-16-20.png filter=lfs diff=lfs merge=lfs -text +references/Screenshot[[:space:]]from[[:space:]]2026-05-19[[:space:]]09-22-23.png filter=lfs diff=lfs merge=lfs -text +references/Screenshot[[:space:]]from[[:space:]]2026-05-21[[:space:]]12-05-26.png filter=lfs diff=lfs merge=lfs -text +references/Screenshot[[:space:]]from[[:space:]]2026-05-29[[:space:]]16-20-02.png filter=lfs diff=lfs merge=lfs -text diff --git a/.gitignore b/.gitignore index 9a4d320cad716e4fcae462eaaf12707155087552..865c3a61ce80efd11ab4ec9e687ae848baf5df8b 100644 --- a/.gitignore +++ b/.gitignore @@ -213,4 +213,10 @@ api/public/ scratch/ docs/ +artifacts/ +data/ + +api/public/ + +logs/ diff --git a/Dockerfile b/Dockerfile index 523c6ca682882aefbff53d13ae4f89d9a4aedd77..85d776c164683b2b6522e85f99a4fcfbffb67371 100644 --- a/Dockerfile +++ b/Dockerfile @@ -1,4 +1,8 @@ FROM python:3.12-slim +sudo apt-get update && sudo apt-get │ │ +│ │ install -y tesseract-ocr │ │ +│ │ libtesseract-dev poppler-utils │ │ +│ │ libmagic-dev WORKDIR /app diff --git a/README.md b/README.md index 2636aada3bbb49f516ede701e4cb13e60aaa19a7..b4c5227789c802d63ef22c5751c65e8a8377d3ae 100644 --- a/README.md +++ b/README.md @@ -9,150 +9,271 @@ pinned: false short_description: This is the Multi-Rag Agent --- +# 🧠 Multi-RAG — Advanced Multi-Modal Retrieval Augmented Generation -
Advanced Multi-Agent RAG Orchestration powered by LangGraph, AWS Bedrock, and FAISS
+
+
+
+
+
+
+
+
+
Built with 💖 for the future of Agentic RAG.
-Knowledge base is ready
+Ask anything about the documents you ingested. The graph-powered RAG will find the answer.
+Enter to send · Shift+Enter for newline
+Loading ingested documents…
++ Upload PDFs, clinical datasets, brain scans, audio logs, or your deepest existential thoughts. + Our AI vectorizes them into neat 384-dimensional math, builds a mind-bending knowledge graph, + and lets you chat with it. + + *Disclaimer: We are not responsible if the LLM starts giving life advice or questioning your queries. + +
+ + + + + +We took the boring parts of Retrieval-Augmented Generation, sprinkled some high-speed agentic loops, and wrapped it in code.
++ Upload clinical CSVs, images, or standard PDFs. Our pipeline ingests and chunks them, ready to feed the vectors. +
++ Powered by all-MiniLM-L6-v2, converting unstructured text into dense math vectors, establishing links instantly. +
++ Query your agent. Using stateful LangGraph chains, the system navigates nodes to fetch context and streams back accurate answers. +
+Your session — and all its data — will be auto-wiped after the timer ends.
+PDFs, CSVs, text, images — any format, any quantity.
+Drag & drop files here
+or click anywhere in this box to browse
+Vectorizing, embedding, building knowledge graph… grab a chai ☕
+Files
+—
+Docs
+—
+Status
+⚙️
+| Layer Type | Complexity per Layer | Sequential Operations | Maximum Path Length |
|---|---|---|---|
| Self-Attention | O(n? - d) | O(1) | O(1) |
| Recurrent | O(n - d?) | O(n) | O(n) |
| Convolutional | O(k-n-d?) | O(1) | O(log, (n)) |
| Self-Attention (restricted) | O(r-n-d) | O(1) | O(n/r) |