If you need to decompress the files, please see the main README at the github repo.
If you want to use them directly from the parquet files, the original .tif/.nc files were read into the rows as binary file data sources
We provide '.csv' files with predefined train (blue), test (orange) and validation (green) splits that can be used for repeatability and comparability of experiments.
60% of tiles are allocated for training, 20% for… See the full description on the dataset page:
https://huggingface.co/datasets/M3LEO/southamerica.
