如何多卡训练 - Githubissues

showlab / Tune-A-Video

[ICCV 2023] Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation

https://tuneavideo.github.io

Apache License 2.0

4.15k stars 377 forks source link

Open chongjg opened 11 months ago

chongjg commented 11 months ago

现在在text-video数据集上进行训练想增大batch size但是一用多卡就会报错，不知道应该如何进行修改呢？

kzl787 commented 5 months ago

我也遇到了这个问题，请问你解决了吗