Hi Csaba,

Good question. Sorry for not explaining that in the article. The reason I changed the deconvolutional layer is that the conv2d_transpose actually creates artifacts that are undesirable. For a more thourough explaination (with visuals) check out this article on it: http://distill.pub/2016/deconv-checkerboard/.

By utilizing depth_to_space, we are able to actually produce more robust and clear images. This technique is sometimes called “subpixel” and you can read about it here:


PhD. Interests include Deep (Reinforcement) Learning, Computational Neuroscience, and Phenomenology.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store