This is a scikit-learn based project which predicts the type of soil based on 10 random samples submitted through a form. The algorithm communicates to the user interface via an API built with FastAPI.
- Create a virtual environment and set a default python version ^3.x:
mkvirtualenv envRefer this if not familiar with python virtual environments
- Activate the virtual environment:
source env/bin/activate- Install all the dependencies:
pip3 install -r requirements.txt-
Clone this project
-
To run and test the project locally, run:
uvicorn main:appthan go to 127.0.0.0:8000/docs to view the swagger UI for the API
The Gunicorn command to deploy the API on cloud has been given in the dockerfile.
However, you can check if it's working properly by running gunicorn -w 4 -k uvicorn.workers.UvicornWorker main:app
The algorithm utilizes scikit-learn, numpy and pandas to generate results. The results are taken to an API using FastAPI and pydantic. To deploy the API on Azure, the backend has been containerized with Docker and shipped with Gunicorn and Uvicorn.
See wix.js, the frontend was created using Wix and it was integrated with the algorithm using Velo. This file contains the fetch code which delivers the algorithm result to the UI.