This dataset contains the data used in the paper "Bridging Logic and Learning: Decoding Temporal Logic Embeddings via Transformers" (Candussio et al. 2025).
More in detail:
train.csv is the training set for the random models. It contains formulae spanning from depth 2 to depth 23;
TODO: complete with test set.
