what's the best cudatoolkit version to use now that is compatible with most of the ai software out t
what's the best cudatoolkit version to use now that is compatible with most of the ai software out there now?
J12.1 or above i think
Ji think most things use 11.8
Jand i have seen some use 12.0
J@Dr. Furkan Gözükara Spam from Hinga.inc
Zi'm a install 11.8
Fwill check messages
Fpreparing kaggle foocus notebook
Lwhat is the fastest way to train reliable checkpoint or lore? i have heared about instant lora and how long does it take?
FGetting super fast speed on sdxl on 4090
FDepends SDXL or SD1.5? for me around 10-15 mins for sd.15 and 30-50 mins on sdxl
Linstant lora just uses ipadapter, it's not actually lora related at all and doesnt yield nearly as good results
Lthats too much
Lyeah i know i saw another video with semilia name , i hoped there would be something faster
Lu know anything faster than that?
Fbest bet is roop or one of those face swappers but quality isnt the best
Lyeah i also want it for animals, pets
Fwhen do you get this error?
Fi am yet to test
Fit is doable. but procudes same pose. i am teaching this with private consultation
Fhopefully today
Fright now making a notebook for foocus on kaggle
Fworking great
Fmostly cuda 11.8
Fcheckpoint best
Fif you need lora extract it
Fwow nice. on windows?
Fyep windows, testing on linux soon
Fnew free Kaggle notebook for Fooocus added : https://www.patreon.com/collection/258108?view=expanded
Hthat means if I keep clicking 'continue training' it will try to continue caching the training dataset from where it left off and eventually complete? (or does it cache from the beggining again?)
is there a way to continue training a checkpoint in Kaggle with more input data after it has been trained once?
is there a way to continue training a checkpoint in Kaggle with more input data after it has been trained once?
Fit starts where it left
Fyes you can upload that checkpoint to kaggle temp and give its path
Fto continue
M@Dr. Furkan Gözükarain kohya version 22.2.1 they made a change to include the unet and text encoders so you didn't have to put it in the extra parameters area. In the json file it shows four training parameters now Do you know what the second text encoder "te" is for: "learning_rate": 1e-05,
"learning_rate_te": 1e-05,
"learning_rate_te1": 3e-06,
"learning_rate_te2": 0.0,
"learning_rate_te": 1e-05,
"learning_rate_te1": 3e-06,
"learning_rate_te2": 0.0,
Fyou can use gui to set each one
Flatest config updated for that
Mgui only shows unet and te1 and te2
Mjust curious what "learning_rate_te": 1e-05" is since I assume the first learning rate is the unet
MI'll try with above config and see what happens.
Mit must be a mistake in the json file since the print training cmd does not show it: --learning_rate_te1="3e-06" --learning_rate_te2="0.0" --learning_rate="1e-05"
Fuse patreon config
Fpreparing magic animate installer for runpod and linux
Mthx - I do use your config. It produces stunning images that fool even my family. Truly amazing
VI’m going for the record. 150fps of a cat with a hat. https://x.com/cumulo_autumn/status/1732309219041571163?s=46
VGuessing sd turbo, with tensor rt
Fdmn u got huge views
Flol even stability ai liked
VOh no that’s not me. Lol. But that user has the same spec machine I do.
Flol i thought it was you