此嵌入專注於修正手指數量與形狀問題，可能導致模型風格改變
v5:
negative_hand-neg訓練於10張圖像而成。
v22
使用75個vector，訓練資料使用v5與其他負面嵌入產生的兩百張圖像。
v49
使用32個vector，訓練資料使用不同風格模型與其他負面嵌入產生的六百張圖像。
Translated by ChatGPT
This embedding focuses on correcting issues related to the number and shape of fingers, which may result in a change in the model's style.
v5:
negative_hand-neg is trained on 10 images.
v22:
Utilizing 75 vectors, the training data consists of 200 images generated from v5 and other negative embeddings.
v49
Utilizing 32 vectors, the training data consists of 600 images generated from different style models and other negative embeddings.
Description
The training data is generated from the old version (v5) and other negative embeddings, totaling 200 images with dimensions of 512*512.
FAQ
Comments (4)
Do you have a link to the model by any chance?
This is a private model, and it may be released in the future.
The new embedding is 10 times larger than the original. What have you changed?
The training data is generated from old versions and other negative embeddings, comprising a total of 200 images with dimensions of 512*512.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.