SciStyle_subset / README.md
nDimensional
Update README.md
33d6ff3
|
raw
history blame
2.39 kB
---
license: creativeml-openrail-m
library_name: diffusers
tags:
- text-to-image
- StableDiffusion
- Stable-Diffusion
- image-generation
- SD1.5
- SDv1.5
- image-generation
---
<style>
img {
max-width: 100%;
height: auto;
}
figcaption {
font-style: italic;
color: white;
padding: 2px;
text-align: center;
}
</style>
---
<div cclass="title" align="center">
<font size="12" color="Cyan"><span style="font-family:Arial">SciStyle</span></font>
</div>
<figure>
<img src="https://huggingface.co./Schisim/SciStyle_subset/resolve/main/images/SciStyle-v2_Grid.jpg?download=true" alt="v2 Grid" width=2048/>
<figcaption>Images generated with SciStyle_v2</figcaption>
</figure>
---
<div class="v1Info" style='font-family:Monospace'>
<font size="4" color="red"><b>V1</b></font><br>
<b>Base Model</b>: Stable Diffusion v1.5<br>
<b>Type</b>: Fine-Tune<br>
<b>Clip Skip</b>: 1<br>
<b>Medium</b>: MultiMedium (Anime, Painting, Digital Art, 3D Render, Comic Illustration)<br>
<b>Caption Style</b>: Natural Language + Booru style.<br>
<b>Dataset Size</b>: Subset, ~1k images<br>
</div>
---
<div class="v2Info" style='font-family:Monospace'>
<font size="4" color="red"><b>V2</b></font><br>
<b>Base Model</b>: Stable Diffusion v1.5<br>
<b>Type</b>: Fine-Tune<br>
<b>Clip Skip</b>: 1<br>
<b>Medium</b>: MultiMedium<br>
<b>Caption Style</b>: Natural Language + Booru Style<br>
<b>Dataset Size</b>: ~4k images<br>
</div>
<div>
<font size="4"><i>What's new in v2?</i></font>
<ul>
<li>More races/species from various Sci-fi and Fantasy universes.</li>
<li>Started fine-tuning photo manipulation into the model for more abstract generations. This idea has since been abandoned and will be moved to its own dedicated photo manipulation model.</li>
</ul>
</div>
----
<div class="v3Info" style='font-family:Monospace'>
<font size="4" color="red"><b>Science n' Dungeons</b></font><br>
<b>Base Model</b>: Stable Diffusion v1.5<br>
<b>Type</b>: Fine-Tune<br>
<b>Clip Skip</b>: 1<br>
<b>Medium</b>: MultiMedium<br>
<b>Caption Style</b>: Natural Language + Booru Style<br>
<b>Dataset Size</b>: ~4.5k images<br>
</div>
<div>
<font size="4"><i>What's new in Science n' Dungeons?</i></font>
<ul>
<li>v2 with fine-tune on DnD race/monsters dataset</li>
<li><b>Note</b>: Less multimedium than v2</li>
</ul>
</div>