Joshua Carney

icon indicating a website link https://jkcarney.github.io/

icon company JHUAPL icon location Laurel, MD icon location software engineer who supports CS & AI/ML research @JHUAPL

Results 4 comments of Joshua Carney

[Dreambooth] Multi-GPU training with accelerate is magnitudes slower than single GPU (non-flax)

That does help. I just want to confirm this before I test myself and verify it: So basically, if I was doing 800 steps on one GPU, I should get...

[Dreambooth] Multi-GPU training with accelerate is magnitudes slower than single GPU (non-flax)

Ok, here's the outcome of my experiment. I used stable diffusion 1.5 as a base with a subject dataset of my face. The single GPU run was done for 900...

[Dreambooth] Multi-GPU training with accelerate is magnitudes slower than single GPU (non-flax)

`accelerate==0.15.0` Not using xformers at the moment.

[Dreambooth] Multi-GPU training with accelerate is magnitudes slower than single GPU (non-flax)

Interesting. I'm not quite sure how exactly the machine is configured, very possible that it's some issue there. I appreciate the assistance though. I'll close this issue since it's more...