Introduction

Contains two programs, one for training the model (PlantAITrainer) and one for using the model (PlantAIPredictor).

PIP's

pip install tensorflow numpy opencv-python scikit-learn matplotlib

Key Sections of the Program

1. Loading the Dataset:
   • ImageDataGenerator is used for loading and preprocessing the images, including splitting the dataset into training and validation subsets.
2. Building the Model:
   • A basic convolutional neural network (CNN) architecture is used with multiple Conv2D and MaxPooling2D layers followed by a Dense layer for classification.
3. Training the Model:
   • The model is trained using the loaded dataset, with accuracy and loss tracked during training and validation.
4. Plotting the Results:
   • The training and validation accuracy/loss are visualized using matplotlib.
5. Saving the Model:
   • After training, the model is saved as plant_recognition_model.keras for future use.

Credits

ChatGPT 3.5 ;-)