README.md

metadata

title: Plant Disease Detection 0.89
emoji: 🌿
colorFrom: green
colorTo: yellow
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false

https://www.kaggle.com/code/walidchaib02/apples-cnn-on-tpu-96/notebook

Plant Disease Detection using CNN

This project uses a Convolutional Neural Network (CNN) trained on a plant disease dataset (e.g., PlantVillage) to classify healthy and diseased plant leaves from images.

Model

• Architecture: Custom CNN with 4 convolutional blocks, followed by dense layers and dropout.
• Input size: 256x256 RGB images.
• Output: Probability distribution over N disease classes.

How to use

1. Upload an image of a plant leaf.
2. The model will predict the top-3 most likely diseases with confidence scores.
3. The interface is built with Gradio.

Dataset

The model was trained on a subset of the PlantVillage dataset (or your custom dataset). It includes N classes covering various crops and diseases.

Files

• best_model.keras – trained model weights.
• app.py – Gradio application.
• requirements.txt – dependencies.

Performance

Validation accuracy: ~XX% (replace with your final accuracy)

Acknowledgements

• Dataset source: ...
• Built with TensorFlow and Gradio.