search

ZichengDuan / TheChosenOne

Unofficial implementation of the paper "The Chosen One: Consistent Characters in Text-to-Image Diffusion Models"
https://arxiv.org/abs/2311.10093
237 stars 22 forks source link

Time For training process #8

Closed Deaddawn closed 8 months ago

Deaddawn commented 8 months ago

Hi, there. Wonderful work! I am now running the training process, but it seems to take a lot of time, wondering what should be a normal duration time for the training process when training on a single v100, say using the default config provided in the config file.

ZichengDuan commented 8 months ago

Thanks for your interest in my code. The training process does take some time since we need to generate 128 images in each loop. But I think it should be done within 3-4 hours on a single v100 card. Since I only used a single 4090 for all the training.

Deaddawn commented 8 months ago

Thanks for your interest in my code. The training process does take some time since we need to generate 128 images in each loop. But I think it should be done within 3-4 hours on a single v100 card. Since I only used a single 4090 for all the training.

Got it ! Thanks a lot. BTW, will multi gpus training supported?