This project implements a feature on a web page that identifies whether crops are infected with diseases or pests. All computations are performed within the browser. The model uses ShuffleNetV2, and the frontend is built with React.

This is an independent project developed to address the need for accessible and efficient crop disease identification.

• Model Training: PyTorch was used for training the model.
• Browser Compatibility: ONNX (Open Neural Network Exchange) was utilized to enable the model to operate in a browser environment.
• Frontend Development: The frontend is designed with HTML and CSS.

To facilitate the model's operation in a browser, it was converted to the ONNX format. This conversion significantly simplifies the deployment of the model outside of a Python environment, enabling direct inference on the browser.

Live Application:https://rice-disease-classify.vercel.app/