Update README.md

README.md

@@ -1,3 +1,61 @@
----
-license: cc
----
+---
+license: cc
+---
+---
+license: cc
+---
+<style>
+    td pre {
+        white-space: pre-wrap;
+    }
+
+    table {
+        margin-top: 0;
+        margin-bottom: 0;
+    }
+</style>
+
+<table style="border-collapse: separate; border-spacing: 10px; width:100%">
+  <tr>
+    <td style="vertical-align: top">
+      <img src="./Examples/01.png" alt="SchoolGirl01" width="100%"/>
+    </td>
+    <td style="vertical-align: top;">
+      <img src="./Examples/02.png" alt="SchoolGirl02" width="100%"/>
+    </td>
+  </tr>
+</table>
+
+# Overview
+A DORA for [Animagine XL V3.1](https://civitai.com/models/260267/animagine-xl-v31). Trained using [OneTrainer](https://github.com/Nerogar/OneTrainer).
+
+**Note**: Since there aren't many avaliable images of Rina, I generated synthetic data with my previous models to expand the dataset. I uplaoded the models that were used to generate the synthetic data, in case you want to try them instead.
+
+
+# Usage
+- **Trigger keywords**: 1girl, arisugawa rina
+- **Recommended keywords**: blue eyes, blonde hair, twintails, blunt bangs
+- **Recommended strength**: 0.5 ~ 0.7
+
+# Examples
+**Note**: All examples in the repo were produced with [AUTOMATIC1111](https://github.com/AUTOMATIC1111/stable-diffusion-webui) without any edits. You download them and load all their settings from the metadata to reproduce similar results.
+
+Highres fix details:
+<pre>Denoising strength: 0.3, Hires upscale: 1.5, Hires upscaler: 4x_NMKD-YandereNeoXL_200k</pre>
+
+<table style="border-collapse: separate; border-spacing: 10px; width:100%;">
+  <tr>
+    <td style="vertical-align: top">
+      <img src="./Examples/03.png" width="100%"/>
+    </td>
+    
+  <td style="vertical-align: top;">
+      <img src="./Examples/04.png"  width="100%"/>
+    </td>
+
+  <td style="vertical-align: top;">
+      <img src="./Examples/05.png"width="100%"/>
+    </td>
+    
+  </tr>
+</table>