MAGNet: Modern Art Generator using Deep Neural Networks

How I used CLIP, SWAGAN, and a genetic algorithm to create modern paintings from text descriptions

By Robert. A Gonsalves

You can see my article here on Medium.

The source code and generated images are released under the CC BY-SA license.

StyleGAN2 and SWAGAN implementations are by rosinality, https://github.com/rosinality/stylegan2-pytorch
Native ops are by orpatashnik, https://github.com/orpatashnik/StyleCLIP

Name		Name	Last commit message	Last commit date
Latest commit History 27 Commits
op		op
1_MAGnet_Gather_Modern_Paintings.ipynb		1_MAGnet_Gather_Modern_Paintings.ipynb
2_MAGnet_Process_Modern_Paintings.ipynb		2_MAGnet_Process_Modern_Paintings.ipynb
3_MAGnet_Filter_Modern_Paintings.ipynb		3_MAGnet_Filter_Modern_Paintings.ipynb
4_MAGnet_Modern_Paintings_Train.ipynb		4_MAGnet_Modern_Paintings_Train.ipynb
5_MAGnet_Generate_Modern_Paintings.ipynb		5_MAGnet_Generate_Modern_Paintings.ipynb
LICENSE		LICENSE
MAGneg2_small.jpg		MAGneg2_small.jpg
README.md		README.md
stylegan2.py		stylegan2.py
swagan.py		swagan.py