The OPS-SAT case: test dataset

The OPS-SAT case: test dataset is the test split of the official dataset of ESA's Kelvins "the OPS-SAT case" challenge, created in collaboration with ESA's Φ-lab and ESA's OPS-SAT spacecraft operations team. It consists of 588 raw patches generated by cropping original acquisitions taken by the OPS-SAT cube-sat using its on-board camera. The patches have a resolution of 200x200 and are provided in .png format. Each patch was assigned through manual labeling to one of the following landcover classes: Agricultural, Cloud, Mountain, Natural, River, Sea_Ice, Snow, and Water.
The competition's goal is, given a model of a neural network (in this case the EfficientNet-Lite0), to provide best possible network parameters to perform an on-board classification task. To this goal, competitors were free to use any data, including The OPS-SAT case dataset, which was released before the start of the competition.
The OPS-SAT case: test dataset was hidden from the competitors during the competition and was used to evaluate and rank their submissions. 
More details about the competition setup and solution evaluation are on the Kelvins competition platform.