This project applies Convolutional Neural Networks (CNNs) to classify different rice varieties based on image datasets.
The model is trained using TensorFlow/Keras, and it supports prediction with visual outputs such as class probabilities, training graphs, and performance metrics.

Rice varieties included:

• Arborio
• Basmati
• Ipsala
• Jasmine
• Karacadag

This project uses the 75K Rice Image Dataset containing 15,000 images for each of the 5 rice classes.

• Source: 75K Rice Image Dataset on Kaggle
• Each class contains 15,000 images (total 75,000).
• Images are divided into training, validation, and testing sets.

├── Train/
│ ├── Arborio/
│ ├── Basmati/
│ ├── Ipsala/
│ ├── Jasmine/
│ └── Karacadag/
└── Test/
├── Arborio/
├── Basmati/
├── Ipsala/
├── Jasmine/
└── Karacadag/

• Training data is split into 80% Train / 20% Validation.
• Test data is used for final evaluation.

• Python 3.x
• TensorFlow / Keras
• NumPy
• Matplotlib
• scikit-learn