metadata
tags:
- unconditional-image-generation
license: apache-2.0
Model info
fbanime.pkl
StyleGan2 model trained with official StyleGan3. But I modified the code (networks_stylegan2.py and dataset.py) to support non-square resolutions.
FID: 1.4
g_mapping.onnx
onnx format mapping network of fbanime.pkl
g_synthesis.onnx
onnx format synthesis network of fbanime.pkl
encoder.onnx
e4e model trained with encoder4editing-stylegan3. I add support for official StyleGan2 model and change backbone to ResNet-34 in restyle-encoder.
waifu_dect.onnx
YOLOv5 model trained with official YOLOv5
Usage
see demo
Dataset
fbanimehq v2.0