In the experiment of the paper, your pre-train image synthesis GAN network and the image captioning GAN network, respectively, then combine them by end-to-end training. I want to reproduce this process, but it is tough because no documentation or comment is available in this repository. Could you please provide detailed instructions for these codes? Thanks.
In the experiment of the paper, your pre-train image synthesis GAN network and the image captioning GAN network, respectively, then combine them by end-to-end training. I want to reproduce this process, but it is tough because no documentation or comment is available in this repository. Could you please provide detailed instructions for these codes? Thanks.