What' is this:
These LORA's are for adjusting style of Ebara Pony.(It will not work with Pony and Pony-derived models other than Ebara.)
It is not meant to make major changes to the style, but can be used for fine adjustments.
Other style adjustment LORAs fro Ebara:
How to use:
No trigger word is needed.
Recommended weights is 1
It is assumed to be used with positive weights, but it can also be used with negative weights.
How was it made::
These LORA's were created by copy machine learning method, and utilize sd-webui-traintrain.
I would be happy if there is anything here that can be of even a little use.
Description
first release
FAQ
Comments (3)
Im pretty new to loras and im wondering how these types of loras are made without affecting artstyle?
This LORA's is created by copy machine learning method. Please look at this.
https://github.com/hako-mikan/sd-webui-traintrain/blob/main/README.md#difference
Personally, I feel that a considerable portion of functional LORA, after being created, achieves the desired effect, but due to some factors beyond the author's consideration, many of them, under the combined factors of color, lighting, details, etc., end up having a style-like effect, even though that is not what the author intended. Therefore, most of the time, we have to find the parts we want through extensive trial and error, and the extra parts can also be utilized when needed.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.