Project Status: Completed 🎉
A vast majority of the Indian farmers believe in depending on their intuition to decide which crop to sow in a particular season. They find comfort in simply following the ancestral farming patterns and norms without realizing the fact that crop output is circumstantial, depending heavily on the present-day weather and soil conditions.
However, a single farmer cannot be expected to take into account all the innumerable factors that contribute to crop growth before reaching a consensus about which one to grow.
FarmEasy🌾 solves this problem by using a combination of analytics and machine learning so that the farmers can make a more informed decision.
- Easy to use - The user interface of the project is very simple and easy to use. The user just needs to enter the required data and the project will predict the crop type.
- Accurate - The project uses machine learning algorithms to predict the crop type. The accuracy of the model is 99.27%.
- Used Docker - The project is containerized using Docker. The project is also integrated with GitHub Actions for continuous integration and continuous deployment.
- Implementated CI/CD - The project is deployed on AWS using Docker. The project is also integrated with GitHub Actions for continuous integration and continuous deployment.
- Deployed on AWS - The project is deployed on AWS EC2 using Docker.
- Responsive - The project is responsive and can be used on any device.
The dataset used for this project is taken from the Kaggle website. The dataset contains 2200 rows and 8 columns. The dataset contains the following columns:
- N - ratio of Nitrogen content in soil
- P - ratio of Phosphorous content in soil
- K - ratio of Potassium content in soil
- temperature - temperature in degree Celsius
- humidity - relative humidity in %
- ph - ph value of the soil
- rainfall - rainfall in mm
- label - crop type
The machine learning model is built using the following steps:
- Load the dataset
- Applying one-hot encoding to the label column
- Split the dataset into training and testing sets
- Apply feature scaling to the dataset
- Apply different classification algorithms like KNN, SVM, Random Forest and Naive Bayes
- Perform ensemble learning using Voting Classifier
- Save the model
The accuracy of the model is 99.27%.
The backend of the project is built using Flask and is hosted on AWS. The backend is responsible for the following tasks:
- Fetch the data from the frontend
- Perform ensemble learning on the data
- Predict the crop type
- And then return the result to the frontend
The frontend of the project is built using HTML, CSS and JavaScript. The frontend is responsible for the following tasks:
- Taking the input from the user
- Connect to the backend and fetch the result
- Display the result to the user
- Clone the repository
git clone https://github.com/abhikalparya/FarmEasy.git- Install the dependencies
pip install -r requirements.txt- Run the app
python app.py