Code for "Generating by Understanding: Neural Visual Generation with Logical Symbol Groundings" https://arxiv.org/abs/2310.17451

The program works for conducting rule learning on the Mario dataset. PyTorch and Swi-Prolog are required for code running.

Dataset Preparation:

• Preliminary step: download "mario_icons" folder from "https://github.com/EleMisi/VAEL" project, and put it into the "create_dataset" folder.
• The dataset can be constructed with "create_mario_positive.py, create_mario_negative.py" in the "create_dataset" folder.
• These two files are adapted from "https://github.com/EleMisi/VAEL/blob/main/utils/mario_utils/create_mario_dataset.py"
• Dataset saving paths can be modified in the "main" function calls of the two Python files.

Conduct Learning:

• Running "main.py" in the root folder.
• Path configurations need to be set properly based on the parser arguments therein.
• More learning configurations can be set in the "exp_config/rule_learning_config.py" file.