my upcoming video is about this. teach yourself and a style to the base model. that is a very good w
my upcoming video is about this. teach yourself and a style to the base model. that is a very good way to achieve this
Fi recorded a new dreambooth extension install video. currently processing it. hopefully will be published today. now it supports torch version 2 as well. i will also show that
Ftoday i am explaining how to install latest version. will be published in few hours hopefully post processing
Fyes if your graphic card is not supporting bf16, you have to use fp16. i am mentioning this in my tutorials actually
Fmust be a configuration error. pathing folder directory path error
Fhi i recorded a new video for how to install
Ftoday will be published hopefully
WOk thanks. Yes, it generate the class correctly but the training all samples are black haha
Fmust be another error
Fperhaps yaml
Fresolution
F2.1 you are using?
Fwhich model you are using to train,?
Wsd15newvaeproned.ckpt
WI just have one installed
Fwhich one is that
Fwhere did you get it
WFrom your video, mmmm it is a luncher EmpireMediaScience
WIt download and install the model from the luncher
W
Fi see they must have added this later
Fi am using this one
Fit is the official base model
Fbest one for training
WOk let me try
WTo install a new model I only need that file and put it on the "stable-diffusion-webui\models\Stable-diffusion" folder?
Fif it is 2.1 version you need also yaml file
Wok I will try 1.5 first. How 2.1 version works? Thanks a lot.
F2.1 is much harder
Ffirst get success with 1.5
Nthat is interesting! looking forward for your next video 
do you mean mixing two concepts- one person and a style or merging two checkpoints? one trainined on an individual and the other on a style-
do you mean mixing two concepts- one person and a style or merging two checkpoints? one trainined on an individual and the other on a style-Fi am first teaching myself
Fthen training style
Nhttps://huggingface.co/plasmo/woolitize-768sd1-5 <-- this is so sweet 
but a trained image i want to combine with me
but a trained image i want to combine with me
Fthen i think i will teach myself again with further training
Fthat model probably baked
Fso your training on that may fail
Nwhat does baked mean?
Fbaked means model lost flexibility with overtraining
Fi will explain in video hopefully
Fwhy to follow my strategy
Nwell- it works, but the sheeps are much cuter then me..
N
Nsounds interesting (once again)!
HInstall instruction on the discord under notebooks
TAwesome, thanks
WNow I'm geeting men images thank!!!
IYes, in your tutorial it was clear. On Runpod I can't tell which one is supported, so I always run the wizard. Yesterday the wizard was always wrong, for some reason. On my local machine, the wizard gives me nearly the best settings! So my only solution was to train, cancel, see if the images were black or not. It took me hours to find the right settings, but it's time consuming. Then on a different pod, it's the same story again, which means paying for extra hours.
