Data files for testing entropy coding software

This series of raw data contains two types of data:a) synthetically generated data: each file comprises 10^8 bytes, while each byte represents a single symbol drawn from an alphabet with K different symbols. The distribution of theses symbols is either uniform or geometrically (truncated). For K > 256, the files contain 2 bytes per symbol.b) real-life data (3 files): one screen shot 1785 x 1225 px and two prediction-error images of same size.