add ligne claire

README.md

@@ -29,6 +29,7 @@ artoria pendragon \(swimsuit ruler\) \(fate\), ahoge, blonde hair, green eyes, f
 ```
 ![artoria pendragon swimsuit ruler](characters/artoria%20pendragon%20\(swimsuit%20ruler\)/extras/example.png)
 
+
 ## Styles
 ### kaoming
 Based off artist: https://twitter.com/kaoming775
@@ -48,3 +49,18 @@ black hair, blunt bangs, straight hair, long hair
 ![a boy](styles/kaoming/extras/example.png)
 
 
+### ligne claire
+LoRA version of https://huggingface.co/breakcore2/ligne_claire_anime_diffusion
+Does not do abstract style as well as the finetuned model.
+
+[Download .safetensors](https://huggingface.co/breakcore2/loras/tree/main/styles/ligne%20claire)
+
+[Example Images with Prompt Metadata](https://huggingface.co/breakcore2/loras/tree/main/styles/ligne%20claire/extras)
+
+Relevant prompts:
+```
+ligne claire, flat color
+can also add in:
+low contrast, chromatic abberation
+```
+![a girl surrounded by lily flowers](styles/ligne%20claire/extras/example.png)

styles/ligne claire/ligne_claire_anime.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52629ed13187d86e1229812228cc1977e45b24d2eb20971f801a4762a23616d3
+size 151112206

styles/ligne claire/recommended prompts.txt

@@ -0,0 +1,3 @@
+ligne claire, flat color
+can also add in:
+low contrast, chromatic abberation

styles/ligne claire/training notes.md

@@ -0,0 +1,21 @@
+## Training Parameters
+Dataset consisted of:
+```
+377 images repeated 2 times per epoch
+```
+
+Training Parameters:
+```
+$learning_rate = 6e-5
+$lr_warmup_ratio = 0.10
+$train_batch_size = 2
+$num_epochs = 4
+$save_every_n_epochs=1
+$scheduler="constant_with_warmup"
+$network_dim=128
+$alpha=128
+$resolution=768
+```
+
+notes:
+"ligne claire" token was kept using --keep-tokens=1