how to train diffusion model using deepspeed
I’m trying to train an sd3 model with 4×3090, how can I optimize it with deepspeed? I get an error when I use ZeRO stage 3
I sincerely hope to get your help
how to train diffusion model using deepspeed
I’m trying to train an sd3 model with 4×3090, how can I optimize it with deepspeed? I get an error when I use ZeRO stage 3
I sincerely hope to get your help