This post is presented in noteworthy - the journal blog.
Hi everybody! Today, we will continue the series about autoregressive models.
Summary Autoregressive models — PixelCNN Modelling coloured images PixelCNN’s blind spot in the receptive field Fixing the blind spot — Gated PixelCNN Conditional generation with Gated PixelCNN Gated PixelCNN with cropped convolutions Improving performance — PixelCNN++ Improving sampling time — Fast PixelCNN++ Using attention mechanisms — PixelSNAIL Generating Diverse High-Fidelity Images — VQ-VAE 2 For each topic, we implemented the models that are available in this repository.