/Potato-disease-classifier

Potato disease classifier using deep learning

Primary LanguageJupyter Notebook

Potato disease classifier

This project is a machine learning model that classifies potato leaf diseases based on images of potato leaves. The model is built using the TensorFlow library and is trained on a dataset of labeled images.

The model can currently classify three different types of potato leaf diseases: Early Blight, Late Blight, and Healthy. The accuracy of the model is currently at 95%, as evaluated on a separate test dataset.

To use the model, simply input an image of a potato leaf and the model will output a prediction of the type of disease present in the leaf.

Dataset

Dataset The dataset used to train the model is publicly available on Kaggle and can be found here. It contains a total of 16,074 labeled images of potato leaves, including 5,376 images of Early Blight, 5,400 images of Late Blight, and 5,298 images of Healthy leaves.

Dependencies

  • TensorFlow 2.4.0
  • NumPy 1.19.3
  • Matplotlib 3.3.3

How To Use?

Clone repository

  git clone https://github.com/irfanrasheedkc/Potato-disease-classifier

Run FastAPI

  cd api
  python main.py

Run Website

  cd frontend
  npm install --from-lock-json
  npm start