GPU Inference – Alias-Free GAN

originally by duskvirkus

This is a notebook for generating images and latent space walks with Alias-Free GAN on a Colab GPU instance.

Repository: https://github.com/duskvirkus/alias-free-gan

An unofficial version of Alias-Free Generative Adversarial Networks (https://arxiv.org/abs/2106.12423). This repository was heavily based on Kim Seonghyeon’s (rosinality) implementation. The goal of this version is to be maintainable, easy to use, and expand the features of existing implementations. This is built using pytorch and pytorch lightning (a framework that abstracts away much of the hardware specific code).

Example of linear interpolation between two random seeds with noise
Example of linear interpolation between two random seeds

Facebook and Power

“facebook.com and apple.com becoming dominant brokers or
gatekeepers […]. These blog service providers, corporate
social network sites, and online hosting services sites are in ‘control’ of the flow of
information providing access to information with better opportunities to spread
information. These corporate sites could refuse to pass on the communication and
block flows of information.”

From

Countering the Social Ignorance of ‘Social’ Network Analysis and
Data Mining with Ethnography