Welcome to the EuroSAT Land Use Classification challenge!

Your goal is to build a model that, given a satellite image, predicts the land use/land cover (LULC) class it belongs to. These classes include types such as Forest, Water, Urban, and more, representing different landscape features across the globe. Accurate land use classification can aid in environmental monitoring, urban planning, and resource management.

EuroSAT dataset consists of high-resolution satellite images from the Sentinel-2 satellite, classified into 10 distinct LULC categories. While the classes are visually distinct, many satellite images—especially those taken at different times of the year—can look quite similar. Your model must be robust enough to differentiate between these subtle variations, ensuring good generalization.

You are provided with the train/ and test/ directories, along with the dataframes train.csv and test.csv. Your task is to predict the LULC class for the images in test/ and submit your predictions to the leaderboard.