Skip to content

training and evaluating a CNN architecture on the plant disease detection task

License

Notifications You must be signed in to change notification settings

othmane42/plant-disease-detection

Repository files navigation

Plant disease detection

Plant disease is one of the main problems in the agriculture since they cause great damages to agriculture crops by significantly decreasing production. It is then essential to have powerful early dignoses to prevent such losses,the solution proposed in this work was based on CNN (Convolution Neural Network ), which was widely used the last years in the classification tasks for their accurate results in production.

Dataset

This dataset was used to train and evaluate the model.

training

the training and evaluating steps was done using Keras,I took also advantage of the transfer learning technique for a faster training (architecture used : VGG16).

Here is the link to the final checkpoint ( trained weights ).

requirements

pip install -r requirements.txt

About

training and evaluating a CNN architecture on the plant disease detection task

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published