Update README.md

README.md

@@ -2,7 +2,7 @@
 license: creativeml-openrail-m
 ---
 
-Trigger words
+### Trigger words
 
 ```
 Anisphia, Euphyllia, Tilty, OyamaMahiro, OyamaMihari
@@ -14,7 +14,7 @@ For `0324_all_aniscreen_tags`, I accidentally tag all the character images with
 For `0325_aniscreen_fanart_styles`, things are done correctly (anime screenshots tagged as `aniscreen`, fanart tagged as `fanart`).
 
 
-#### Settings
+### Settings
 
 Default settings are
 - loha net dim 8, conv dim 4, alpha 1
@@ -28,17 +28,28 @@ The configuration json files can otherwsie be found in the `config` subdirectori
 However, some experiments concern the effect of tags for which I regenerate the txt file and the difference can not be seen from the configuration file in this case.
 For now this concerns `05tag` for which tags are only used with probability 0.5.
 
-#### Some observations
+### Some observations
 
 For a thorough comparaison please refer to the `generated_samples` folder.
 
+#### Captioning
+
+Dataset, in general, is the most important out of all.
+The common wisdom that we should prune anything that we want to be attach to the trigger word is exactly the way to go for.
+No tags at all is terrible, especially for style training.
+Having all the tags remove the traits from subjects if these tags are not used during sampling (not completely true but more or less the case).
+
+![00066-20230326090858](https://huggingface.co/alea31415/LyCORIS-experiments/resolve/main/generated_samples/00066-20230326090858.png)
+
+
+#### Others
+
 1. I barely see any difference for training at clip skip 1 and 2.
 2. Setting text encoder learning rate to be half of that of unet makes training two times slower while I cannot see how it helps.
 3. The difference between lora, locon, and loha are very subtle.
 4. Training at higher resolution helps generating more complex backgrounds etc, but it is very time-consuming and most of the time it isn't worth it (simpler to just switch base model) unless this is exactly the goal of the lora you're training.
-5. What I see that makes the biggest difference is in captioning. The common wisdom that we should prune anything that we want to be attach to the trigger word is exactly the way to go for. No tags at all is terrible, especially for style training. Having all the tags remove the traits from subjects if these tags are not used during sampling.
 
-#### Datasets
+### Datasets
 
 Here is the composition of the datasets
 ```