I can not find the inference part for generate avi files from a given model, could you also share them in this repo? Thanks a lot.