DRAW

My implementation of "DRAW (Deep Recurrent Attention Writer)" and "VAE(Variational Auto-Encoder)"

This code is able to deal with images with multiple channels

Generate results on MNIST:

References:

[1] https://github.com/ericjang/draw
[2] Doersch, Carl. "Tutorial on variational autoencoders." arXiv preprint arXiv:1606.05908 (2016).