Converter from bin to ckpt works!!!!! heres link with code from Josh to incorporate to colab !!! about fast-stable-diffusion HOT 9 CLOSED

Yes just saw it, I'll add it to the colab

from fast-stable-diffusion.

Awesome! im super excited , i do have colab pro but i rarely get v100 and just once per month a100, mostly p100 not even T4!!!
But hey its on for many hours + i have 10 gmails for free colabs , i will let you know if ill get one of those

from fast-stable-diffusion.

``You think You can prune the ckpt so its 2gb ?
Heres the prune code
https://github.com/JoePenna/Dreambooth-Stable-Diffusion/blob/main/prune_ckpt.py
Converts 12gb to 2gb after dreambooth on vast or runpod

# This version should automatically prune around 10GB from the ckpt file last_checkpoint_file = directory_paths[-1] + "/checkpoints/last.ckpt" !python "prune_ckpt.py" --ckpt {last_checkpoint_file}

from fast-stable-diffusion.

Would also be grateful for attribution, @TheLastBen - if you could just put a link in the converttosd.py file that goes to my github that would be excellent. :)

from fast-stable-diffusion.

Sure, I was planning on adding that on the main Readme, thanks for the script, amazing job !

from fast-stable-diffusion.

if you have colab pro, let me know if the V100 and the A100 work without issues

from fast-stable-diffusion.

I'll add to the dreambooth the prior preservation to avoid messing up all the classes and get better results.

from fast-stable-diffusion.

ok pruning is in the new code now by default ! Amazing

from fast-stable-diffusion.

The CKPT produced by this colab is 3.97 Gb, no need to prune

from fast-stable-diffusion.