adding cluster to EfficientNet project

[TaraStrh]

In this project, I explore a self-supervised approach to analyzing chest X-ray images using EfficientNet, a state-of-the-art convolutional neural network. Instead of traditional supervised classification, we utilize EfficientNet to extract deep image features and apply KMeans clustering to group the images without any label information. This method enables meaningful separation of medical images into categories (e.g., healthy vs. abnormal) even in the absence of annotated data. Our results suggest that EfficientNet-based feature extraction captures essential structural patterns that support effective clustering.