Prakriti Determine

Determining an individual's Prakriti, a fundamental concept in Ayurveda has long been a challenging and intricate process. Traditional Prakriti assessment methods required extensive consultations with Ayurvedic practitioners, making it arduous and time-consuming. Prikriti Determine features a Machine learning-based chatbot capable of efficiently determining a user's Prakriti type. The chatbot assesses various facets of their body and mind by engaging users in a friendly conversation and asking a series of well-crafted questions. This approach allows for a more consistent and objective Prakriti analysis. Based on the Prakriti, the chatbot gives diet recommendations.

Architecture of the Used Model FNN

Watch the Output video 👉 Video

Installation Guide

ChatBot 💻 Installation & Training Guide

1. Make Sure you have installed the Python and version should be >=3.10.

2. Navigate to the bot Folder using command

   cd bot
   

then create the Models Folder inside it

mkdir Models

3. Create Virtual Environment using virtualenv or any package you need here I'm preferring virtualenv

   1. Install the virtualenv package using pip

      pip install virtualenv
      

   2. Create the Virtual Environment

      virtualenv project
      

   3. Activate the Virtual Environment.

      For windows :

      project/Scripts/activate
      

      For Ubuntu :

      source project/bin/activate
      

4. Before installing packages, first download the GTK Runtime in your system. These Runtime Environment will help you to create the PDF from the HTML also the backend of the package weasyprint

   Download the Runtime Environment for Windows from Here GTK Runtime

   Other Operating System refer to the Installation process of the weasyprint package Read here

5. Install the packages required to run the project.

   pip install tensorflow pandas nltk scikit-learn sqlalchemy fastapi uvicorn websockets weasyprint
   

Note : Wait for the packages to be installed if you encountered any error or problems try to install the packages one by one.

6. Train the Chatbot Model on the same terminal. Make sure you are in the bot folder in terminal.

   For Windows:

   python Training/botmodel.py
   

   For Ubuntu:

   python3 Training/botmodel.py
   

7. Train the Prakriti Model on the same Terminal.

   For Windows:

   python Training/prakritimodel.py
   

   For Ubuntu:

   python3 Training/prakritimodel.py
   

8. Run the API to serve both trained Model to connect with the Frontend.

   For Windows:

   python app.py
   

   For Ubuntu:

   python3 app.py
   

9. Wait for the API to startup then copy the url in which the api is running.

   e.g: https://127.0.0.1:8000 only copy part from 127.0.0.1:8000

   The URL may be different for different OS make a note of it.

The API is Running Successfully 😃😃😃😃.

Frontend Installation 🖼️

1. Make sure you have installed Nodejs in your system.

2. Navigate to frontend folder.

cd frontend

3. Install the packages.

npm i

4. After Installing all the packages make some changes in .env file which is located in frontend folder.

VITE_API=ws://127.0.0.1:8000

Paste your copied url or server url in place of 127.0.0.1:8000

5. Run the Frontend.

npm run dev

The frontend is Running Successfully 😃😃😃.

You have successfully setup your project.

Author: Prathamesh Dhande