Resnet18 Eurosat

🚀 Satellite Image Classification Model

This project presents a fine - tuned ResNet - 18 model for satellite image classification on the EuroSAT dataset, offering high - accuracy results.

🚀 Quick Start

This section will guide you through the basic steps to get started with the model. However, specific start - up code is not provided in the original document.

✨ Features

• Based on the ResNet - 18 architecture, suitable for satellite image classification tasks.
• Fine - tuned on the EuroSAT dataset for better performance.

📚 Documentation

Model Details

This model is based on the ResNet - 18 architecture and it has been fine - tuned for satellite image classification tasks on the EuroSAT dataset.

Model Sources

• Repository: https://github.com/chathumal93/EuroSat - RGB - Classifiers

Training Details

Training Data

The dataset comprises JPEG composite chips extracted from Sentinel - 2 satellite imagery, representing the Red, Green, and Blue bands. It encompasses 27,000 labeled and geo - referenced images across 10 Land Use and Land Cover (LULC) classes.

Training Procedure

• Preprocessing: Standard image preprocessing including resizing, center cropping, normalization, and data augmentation techniques [RandomHorizontalFlip and RandomVerticalFlip].

Training Hyperparameters

• Learning rate: 3e - 5
• Batch size: 64
• Optimizer: AdamW
• Scheduler: PolynomialLR
• Loss: CrossEntropyLoss
• Betas = (0.9, 0.999)
• Weight_decay = 0.01
• Epochs: 20

Evaluation

Results

📄 License

This project is licensed under the MIT license.