Arabic Dates Classification is a deep learning project that uses a fine-tuned ResNet50 model to classify different varieties of Arabic dates. It demonstrates the application of transfer learning and computer vision in agricultural automation and quality assessment.
This project applies ResNet50—a deep residual convolutional neural network pretrained on ImageNet—to recognize and classify varieties of Arabic dates based on subtle differences in color, texture, and shape. The training and evaluation process is implemented within a single, interactive Jupyter Notebook.
- Model: ResNet50 (fine-tuned on custom dataset)
- Framework: PyTorch
- Objective: Automate fruit classification for agricultural and commercial applications
- Language: Python 3.8+
- Deep Learning: PyTorch, torchvision
- Data Handling: NumPy, Pandas
- Visualization: Matplotlib, Seaborn
- Environment: Jupyter Notebook / Google Colab
├── LICENSE ├── README.md ├── NOTICE └── arabic-dates-classification-resnet50.ipynb
The entire pipeline — including data preprocessing, model training, validation, and evaluation — is contained in the arabic-dates-classification-resnet50.ipynb notebook.
# Clone the repository
git clone https://github.com/YourUsername/Arabic-Dates-Classification.git
cd Arabic-Dates-Classification
# Launch Jupyter Notebook
jupyter notebook arabic-dates-classification-resnet50.ipynb - PyTorch for providing the deep learning framework.
- Open-source agricultural image datasets and research communities for reference data.
- Inspiration from practical applications of computer vision in sustainable farming.
This project is licensed under the Apache-2.0 License.
Feel free to use, modify, and distribute with proper attribution.