In this project I trained a CNN model and predicted three types of potato leaf. Either the potato may be healthy or has an early blight disease or late blight disase. The model has good accuracy on these 3 classes. But it is accepting only images of size (256,256) if we pass images other than that shape it won't work. As a future work I have planned to insert resizing function. I used FastApi as a back end implementation to call the model and predict the disease.
- Python, HTML, CSS, JavaScript, jQuery
- FastApi
git clone
https://github.com/Micky373/potato_disease_classification_using_deep_learning.git
cd potato_disease_classification_using_deep_learning
pip install -r requirements.txt
cd api
python main.py
Then cd .. to go to the frontend foler
Here run
npm install --from-lock-json
and when it finshesnpm audit fix
After the installation completed run
npm run start
Then drag and drop any images of potato leaf which is a size of (256,256)
Then the result will be shown shortly
The EDA and model training is contained in the notebooks folder the notebook is
EDA_model_training_and_saving.ipynb
Contributions, issues, and feature requests are welcome!
Feel free to check the issues page.
Give a ⭐️ if you like this project!
- Special thanks to Dhaval Patel
- The UI implementation and code flow done with the help of this play list