This lora will not influence checkpoint style.
train data 752x752
0.6-1 for solo twins
more than 1 for 2-head solo
suggested checkpoint: Anything v3-v5，realistic
Negative prompt can be used a certain. Dont use negative prompt to fix bed images
DO NOT use "Solo","1","2","multiple" in prompt, this lora has no trigger words.
Restore faces is ok.
just support front-view.
This lora is experiment, please don't give me low stars
I have tried to train this lora for 5 times, it is hard to generate 2 head for one hundred percent. otherwise, overtrained.
这个lora我回炉练了4，5遍，过拟合的时候一定出现两个头，但是画面不能看，目前这个版本是最佳了，用1的权重，2-4 batch size 出两张以上双头。
Please give me 5 stars and like 😍, which is important to me.
🎉your reviews motivate me to update，your suggestion for updates are also helpful！
Hope any feedback from you, you can link me on discord, huggingface, github.
All train data collection are human-manual, and training cloud environment costs me much. hope your donation to support my new model. https://ko-fi.com/cecily34
If you have any questions related to the model or its samples, how to get started or just want to share your creations, feel free to join my discord: https://discord.gg/GgZFF3qPEB
That is my pleasure.
🌸 Loves U.
Description
FAQ
Comments (6)
Can this work in a negative prompt?
or would it have to be negative embedding for that?
badhands.pt is available for this model, other negative embedding to improving quality is also suitable. Because if you set this lora (0.5-0.8), you may generate 2 twins, in this case, it is free to use any negative embedding. if you want to generate 2hand solo, please remove deep negative embedding. Hope it help.
I only get this error message:
NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check. Time taken: 0.19sTorch active/reserved: 2272/2368 MiB, Sys VRAM: 4872/20470 MiB (23.8%)
This error often occurs. Do you try to "the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this."?
actually, I just use lora weight 0.9 to fix this error, batch size more than 2, you can try if it works.
The detail information is in github
(There is a rather long ramble. Have a cup of coffee in hand.)
Being born human, I want to create creatures with my own hands. This is the LORA for you. I do not recommend it otherwise.
If you use LORA in the normal way, there is a high probability that SD will refuse to generate creatures, as if to say, "The world doesn't need creatures.
Here is a successful method of going crazy for humans who want to go crazy.
1. Write a prompt as usual. Here, write <lora:shuang:0>.
2. If you generate it here, SD will throw an error. Since <lora:shuang:0> is generating creatures even in a state where we don't know if it is working or not, it is assumed that SD has self-imposed restrictions for the sake of humanity before it goes crazy.
3.So, please select prompt S/R from X/Y/Z plot, which is necessary to make SD go crazy.
4. Fill in <lora:shuang:0>,<lora:shuang:1> in the X axis. Don't forget the [,].
5. Before generating, check the "Include subimages" checkbox at the bottom of the screen (top right of the four checkboxes).
6. Generate again. The creature will have been generated after a decent picture. As you can see, an extra picture will be generated. The extra cost of the creature, one sacrifice is required to put an advanced monster on the field.
7. Love the creature you created. Then post it on CIVITAI. Good Luck.
(かなり長い駄文があります。コーヒーを片手にどうぞ)
人間に生まれたからには、自分の手でクリーチャーを生み出したい。そんなあなたにおすすめのLORAです。それ以外には勧めません。
通常の使い方をすると高確率で「世界にクリーチャーは不要」と言わんばかりにSDが生成を拒否します。SDは人類と違って正気です。
ここからは狂いたい人類向けの成功した狂い方を書いておきます。
1.通常通りpromptを書きます。ここでは<lora:shuang:0>と書いてください。
2.ここで生成するとSDがエラーを吐きます。なお<lora:shuang:0>という機能しているかわからない状態ですらクリーチャーを生成しているので狂う前の人類のためにSDが自主規制しているものと思われます。
3.ですのでおもむろにX/Y/Z plotからprompt S/Rを選択してください。SDを狂わせるための必要事項です。
4.X軸でいいので<lora:shuang:0>,<lora:shuang:1>と記入してください。[,]は忘れずに。
5.生成を！の前に画面下の方の[サブ画像を含める](4つのチェックボックスの右上です)をチェックしてください
6.改めて生成です。まともな絵のあとにクリーチャーが生まれていることでしょう。おわかりと思いますが、余計に絵が生成されます。クリーチャーの追加コスト、上級モンスターを場に出すには1枚生贄が必要なのです。
7.あなたが生み出したクリーチャーを愛でましょう。そしてCIVITAIに投稿しましょう。グッドラック。
text too long, didn't even read. Ruck yo ass