metadata

license: unknown
datasets:
  - ylecun/mnist
pipeline_tag: text-to-image
tags:
  - Stylegan
  - Conditional

Model Card for conditional StyleGAN3-t-MNIST

This model allows to generate MNIST-like images. The model is based on StyleGAN3-t and used the MNIST training split padded to 32x32. The prepared dataset can be found here.

The model was trained using the following parameters:

gamma: 0.125
kimg: 5000
cbase: 16384
batch: 32
metrics: fid50k (3.623 final)

This training took ~48h on 1 Nvidia A100 GPU