Train the encoder + GRU memory on MiniPong, then run the model as the game and watch the dream.
Download this notebook and open it in Google Colab. Work through the exercises.
