Tiny-diffusion: A minimal implementation of probabilistic diffusion models
1–7 of 7 posts
Re: Tiny-diffusion: A minimal implementation of probabilistic diffusion models
#2This is where I see edge ai going. Minimal implementations running on embedded systems everywhere.
Re: Tiny-diffusion: A minimal implementation of probabilistic diffusion models
#3[dead]
Re: Tiny-diffusion: A minimal implementation of probabilistic diffusion models
#4Very cool showing the outputs of hyperparameter search. Hopefully this gets someone out of a tuning tarpit!
Re: Tiny-diffusion: A minimal implementation of probabilistic diffusion models
#5This is very cool. I implemented a slightly more complicated version of this with class guidance a while ago: https://github.com/aestuans/mnist-diffusion
Re: Tiny-diffusion: A minimal implementation of probabilistic diffusion models
#6This is where I see edge ai going. Minimal implementations running on embedded systems everywhere.
Full agree - this is where AI is finding it's main useful niche. Massive models are fun to use but specialization within embedded systems is the clincher.
Re: Tiny-diffusion: A minimal implementation of probabilistic diffusion models
#7This is where I see edge ai going. Minimal implementations running on embedded systems everywhere.
Full agree - this is where AI is finding it's main useful niche. Massive models are fun to use but specialization within embedded systems is the clincher.
Having AI running in every edge location poses an existential threat. They will provide the peripheral sensors for a larger intelligence.