Updated: Sep 10, 2024
characterDownload
1 variant available
The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.
IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.
基本参考这篇文章来训练的：https://civitai.com/articles/7097/flux-complete-lora-settings-and-dataset-guide-post-mortem-of-two-weeks-of-learning
使用了40张图片来训练，2.5e-5的学习率训练了8000步左右，DIM 32，Alpha 16.用joycaption自然语言打标加上关键触发词：Gwenfl6
我感觉Flux的动漫风格基础比较原始，尽量配合其他优秀的Style lora使用会更好
我觉得我训练Lora的方法还有提升的空间，如果有建议请告诉我
Basically refer to this article for training: https://civitai.com/articles/7097/flux-complete-lora-settings-and-dataset-guide-post-mortem-of-two-weeks-of-learning
40 images were used for training, and the learning rate of 2.5e-5 was trained for about 8,000 steps, DIM 32, Alpha 16. Joycaption natural language labeling was used, and the key trigger word was added: Gwenfl6
I feel that the anime style of Flux is relatively primitive, and it would be better to use it with other excellent Style lora
I think there is room for improvement in my training method for Lora. If you have any suggestions, please let me know.