InnerSight is a sentiment analysis LLM Integrated - Web App for Therapeutic Conversations.
It Integrates LLMware library
to perform Retrieval Augmented Generation (RAG) based document analysis on collection of medical, mental illness & therapeutical books
providing sentiment analysis and a QA Chatbot for Conversations with the LLM used
llmrails/ember-v1
: Embeddings Model for generating text embeddings. LinkTheBloke/zephyr-7B-beta-GGUF
: Used as the Generative LLM Model
LLMWARE
: AI Toolkit - Document, Text Chunking, Embeddings & Generative ModelFastAPI
: Backend Web Server for Serving LLMReact.js + Axios
: Client Side Web App
- First Clone the repo
git clone https://github.com/ShubhamTiwary914/innersightLLM.git
cd innersightLLM
- Install Dependencies
Server Side - Create Virtual env(conda) and install fastapi dependencies
cd api
conda create --name innersight
conda activate
pip install -r requirements.txt
uvicorn server:app --port 8000
Client Side - React App
cd app
npm install
npm run dev
- After successfull, your app should be running at: http://localhost:3100/