Example of the uniform encoding grid of N² rectangles (with N = 30) used to encode the planets over the logarithm of the mass and semi-major axis of the planets. For each rectangle, we attribute a unique Unicode character, in order to encode a planet lying inside. In blue the density distribution is shown for the population of the 25 000 planetary systems used to train the model. The levels correspond to fractions of 0.0001, 0.001, 0.01, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, and 1. Note that if two planets (from the same system) are located in the same rectangle, they are encoded as two identical characters.