ravimashru / iqas

Automated FAQ Bot generation through Knowledge Source

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

iqas

Automated FAQ Bot generation through Knowledge Source

Components:

  • Model: Contains dockerized Inference model accessible over REST API.
  • Server: Contains the impelentation of API which end-users will consume. //to-do swagger
  • Admin-client: Contains the UI for the admin (managing Knowledge bases)
  • Client: Contains UI for end-users(querying for answers)

Installation

Server

  • Run mongodb in Docker sh docker run -d -p 27017:27017 --name iqas-mongo -e MONGO_INITDB_ROOT_USERNAME=mongoadmin -e MONGO_INITDB_ROOT_PASSWORD=secret mongo
  • Open IntelliJ or Eclipse
  • New Project -> Existing Maven Project from VCS -> point to this repo
  • Assuming the above steps works
  • Maven dependency installation will start in backgroud, might take a while
  • Navigate to 'ServerApplication.java' and run as 'spring application'

Building documentation

Install mkdocs.

Run npm run build:docs to build the site and add the swagger spec.

Model

  • Go into the model directory
  • Create and activate a Python virtual environment (optional but recommended)
  • Install dependencies: pip install -r requirements.txt
  • Go into the server directory and run uvicorn main:app --reload

About

Automated FAQ Bot generation through Knowledge Source


Languages

Language:Jupyter Notebook 59.7%Language:HTML 21.9%Language:Java 17.6%Language:Python 0.7%