YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
nouns d-fusion
Stable diffusion model trained on Noun avatars
About
With stable diffusion still in its early days finding the perfect spot often requires trial and error approach. During our research we've explored numerous training techniques, datas
ets and parameters.
Current version of the model (v1) was trained using diffusers textual inversion method.
Dataset
Provided model was trained on dataset consisting of first 300 Noun avatars (tokens 0-299).
For better composability every avatar included in the dataset have had background color cha
nged to black before being resized down to 256x256 pixels and rasterized.
- Downloads last month
- 5