Skip to content

Latest commit

 

History

History
66 lines (46 loc) · 3.06 KB

README.md

File metadata and controls

66 lines (46 loc) · 3.06 KB
Raw

👩‍🌾🌿 CropFusionAI

🔥 Official Website 👉 https://cropfusionai.vercel.app

An Open source Crop and Fertilizer Recommendation tool for Farmers. Machine Learning enabled system that recommends the best crop to grow from across 22 different classes of crops based on various metrics including soil type, rainfall, humidity, and nutrient levels.

🎯 Objective

The objective of this project is to use AI and open source technology to help poor farmers in India grow the best crops and fertilize like a pro. Our tool provides reliable and accessible recommendations based on local soil and weather conditions, helping farmers optimize their crop yields and improve their economic livelihoods.

By using open source technologies, we also hope to foster a community of farmers and developers who can collaborate and continuously improve the accuracy and utility of the tool. Let's use the power of AI and open source principles to empower poor farmers in India and beyond!

Project Workflow

Below are some of the stages of this project for sucessfull development and deployment of both frontend and backend services.

  • Data Collection & ML model training
    • In this stage we collect the required training/testing datasets for building the crop & fertilizer recommendation models. Find training notebook here.
  • Backend Deployment of ML model API's
    • Once we have build the ML models we deploy and expose them as API. In our case, we have opted for FastAPI for building the backend services here
  • Frontend Interface with 3D model
    • Once the backend has been setup , we create a user friendly frontend application with ReactJS, so that even non-technical people can easily interact with the ML models.

👩‍💻To Run (Locally)

  1. Git clone the project repository on your local system
git clone https://github.com/deepeshdm/CropFusionAI.git
cd CropFusionAI
  1. Install dependencies in package.json
npm install
  1. Deploy project on local server
npm start

Improvements to make

  • Optimizing the 3D model to reduce loading time.
  • Training the ML models on a larger dataset to provide generalized results.
  • Adding cache support to speed up output time.

Links to Resources

  • Backend Code for Rest API's here
  • Crop Recommendation Dataset here
  • Fertilizer Recommendation Dataset here
  • 3D Model here