Uploaded v2.0.

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.avif filter=lfs diff=lfs merge=lfs -text

README.md

@@ -7,17 +7,23 @@ base_model:
 pipeline_tag: text-to-image
 tags:
 - art
+datasets:
+- v2ray/mike-no-hito
+library_name: peft
 ---
 # NoobAI XL LoRA Mike No Hito
 This is a LoRA for the [v0.5 version of the NoobAI XL model](https://civitai.com/models/833294?modelVersionId=968495), the v1.0 and v0.75 is actually worse than the v0.5 version thus this LoRA isn't trained for those 2 newer versions. Please use the older v0.5 version for the optimal experience.
 
-The dataset used to train this LoRA is scraped using the tag `mike_no_hito` from [Gelbooru](https://gelbooru.com), containing a total of 22 images. The script used to scrape can be found in [v2ray/anime-collection](https://huggingface.co/datasets/v2ray/anime-collection).
+The dataset [v2ray/mike-no-hito](https://huggingface.co/datasets/v2ray/mike-no-hito) used to train this LoRA is scraped using [LagPixelLOL/mitgw](https://github.com/LagPixelLOL/mitgw), containing a total of 172 images.
 
-Big thanks to the artist for the very cute catgirls :3, though the X (Twitter) link to the artist seems to be broken.
+Big thanks to the artist for the very cute catgirls :3, you can find the artist on X (Twitter) with ID [@doremifaso64](https://x.com/doremifaso64).
 
-To use this LoRA, you can use the trigger word `mike no hito`. The artist is a furry artist so this will basically turn any character into a furry, it also seems unable to make any character without animal ears. So please take in mind these 2 limitations.
+To use this LoRA, you can use the trigger word `mike no hito`, it's best to add `black eyes` and `white pupils` too for better eye performance. The artist is a furry artist so this will have a chance to make a character into a furry unprompted, if you encounter this, simply add `furry` to the negative prompt.
 
-This LoRA is trained using [kohya-ss/sd-scripts](https://github.com/kohya-ss/sd-scripts), with rank 8, alpha 16, learning rate 1e-4, batch size 22 since there are only 22 images, for 256 epochs, using a H100, took approximately 30 minutes.
+This LoRA is trained using [kohya-ss/sd-scripts](https://github.com/kohya-ss/sd-scripts), with rank 8, alpha 16, learning rate 1e-4, for 64 epochs with a total of 1216 steps, using a H100, took approximately 1 hour.
+
+## v2.0 Changelog
+This version added a lot more training data, which are hand picked, cleaned, and deduped by me, mostly by hand. It has improved non-furry ability thanks to generalization, and overall better quality.
 
 ## Examples
 ![](https://huggingface.co/v2ray/nai-lora-mike-no-hito/resolve/main/examples/0.avif)

mike_no_hito.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af0ef26268cefd9f5f4c7b6be99f73789004bb08aa5ea3ab1555558faad047e9
-size 42884100
+oid sha256:29bfdb09688ec85103aa1681b93588168c537fe4bf87fef867edd7d5038a4c7f
+size 42902796