YoungMasterFromSect commited on
Commit
d050cde
1 Parent(s): e8a5daf

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -0
README.md CHANGED
@@ -23,6 +23,11 @@ All of my LoRAs are to be used with their original danbooru tag. For example:
23
  ```
24
  asuna \(blue archive\)
25
  ```
 
 
 
 
 
26
  The usage of them is also explained in this guide:
27
  https://github.com/kohya-ss/sd-webui-additional-networks#how-to-use
28
 
 
23
  ```
24
  asuna \(blue archive\)
25
  ```
26
+ My LoRAs will have sufixes that will tell you how much they were trained. Either by using words like "soft" and "hard",
27
+ where soft stands for lower amount of training and hard for higher amount of training,
28
+ or by Epochs, in short higher epochs = more training
29
+ More trained LoRA is harder to modify but provides higher consistency in details and original outfits,
30
+ while lower trained one will be more flexible, but may get details wrong.
31
  The usage of them is also explained in this guide:
32
  https://github.com/kohya-ss/sd-webui-additional-networks#how-to-use
33