Update README.md

README.md

@@ -5,6 +5,15 @@ tags:
 - diffusers
 - template:diffusion-lora
 widget:
+- text: '-'
+  output:
+    url: images/ComfyUI_00789_.png
+- text: '-'
+  output:
+    url: images/ComfyUI_00796_.png
+- text: '-'
+  output:
+    url: images/ComfyUI_00793_.png
 - text: >-
     an anime illustration of kitsune, girl, blue eyes, braided hair,
     multicoloured hair, brown hair, pink hair, brown fox ears, brown fox tail,
@@ -14,15 +23,6 @@ widget:
     storytelling.
   output:
     url: images/ComfyUI_00784_.png
-- text: '-'
-  output:
-    url: images/ComfyUI_00789_.png
-- text: '-'
-  output:
-    url: images/ComfyUI_00793_.png
-- text: '-'
-  output:
-    url: images/ComfyUI_00796_.png
 base_model: tencent/HunyuanVideo
 instance_prompt: an anime illustration of
 license: mit
@@ -54,9 +54,14 @@ Use the following resolution for the best results:
 
 ## Training
 
-The model has been trained on a tag-balanced dataset of 2k best pixiv illustrations, at resolution of 768, for 856 eopchs. (214e * repeat4)
 
-The training takes about 4 days on a 8xH100 cluster.
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/636982a164aad59d4d42714b/oMsfEYPYbWLyK6mihIkWm.png)
+
+The model has been trained on a tag-balanced dataset of 2k best pixiv illustrations, at resolution of 768, for 856 eopchs (214 epochs * 4 repeats per epoch). 
+
+
+The training takes about 4 days on a 8 x H100 cluster. By the time training ends the loss is still consistently going down, so further training could be beneficial.
+