Rose

I'm going to use the 4090 to fine-tune the large model of SDXL but I keep saying cuda is insufficient when I drive bisz to 1.

2

running training / 学習開始 num examples / サンプル数: 6420 num batches per epoch / 1epochのバッチ数: 6420 num epochs / epoch数: 1 batch size per device / バッチサイズ: 1 gradient accumulation...

Multi GPU training checkpoint problem

1

 I am currently using two 4090s to train sdxl checkpoint, but the speed does not seem to increase and the epoch I set has changed from 10 to 20....

现在好像有个新的问题，前面下载了一些其他的笔记后面在下载不同的会提示数据获取失败

1

希望作者能够修复或者优化下这个

Train FLux checkpoint Lack google/t5-v1_1-xxl Help Me guys！

1

I want to know which files need to be downloaded for the google/t5-v1_1-xxl FLux training script? There are two super-large 44.5G files under google/t5-v1_1-xxl, which makes my download speed very...

采集搜索数据找不到storage format 文件

2

storage format 找不到