My stable diffusion model doesnt produce any images anymore
My stable diffusion model doesnt produce any images anymore
Dits all black when I want to produce a sample image
DHow can I fix this?
FNo idea what did you do so can't tell :/
FDo fresh install always best solution
DWon't I lose my model this way?
FAre you able to generate images with other models?
DI am not
DIs it a dreambooth issue then or automatic1111?
MI trained a LORA model with as base model 1,5pruned.
As class images i used Photo of Man
and i had 40 images of my self as training images called Quirijn
my sample prompt was photo of quirijn man
After 150 epochs the sample images looked really good.
I stopped the training and went to txt2img
As model i selected the most recent savetensors model from my training
When i generate an image with the prompt quirijn man, it looks really good,
but the prompt "an apple on a table" is not working, it only remembers quirijn man.
How can i fix this, should i merge the model with 1,5pruned?
As class images i used Photo of Man
and i had 40 images of my self as training images called Quirijn
my sample prompt was photo of quirijn man
After 150 epochs the sample images looked really good.
I stopped the training and went to txt2img
As model i selected the most recent savetensors model from my training
When i generate an image with the prompt quirijn man, it looks really good,
but the prompt "an apple on a table" is not working, it only remembers quirijn man.
How can i fix this, should i merge the model with 1,5pruned?
FYou can save backup of your model always
THas anyone tried merging an inpainting model with a custom dreambooth model? Is it possible?
TI read on Reddit that you can do it by setting model A = inpainting model , model B = custom model , model = base 1.5 model
TThen merge as difference with strength = 1
THowever I get: RuntimeError: Error(s) in loading state_dict for LatentDiffusion:
size mismatch for model.diffusion_model.input_blocks.0.0.weight: copying a param with shape torch.Size([320, 9, 3, 3]) from checkpoint, the shape in current model is torch.Size([320, 4, 3, 3]).
size mismatch for model.diffusion_model.input_blocks.0.0.weight: copying a param with shape torch.Size([320, 9, 3, 3]) from checkpoint, the shape in current model is torch.Size([320, 4, 3, 3]).
Twhen I tried to load this model
Fthat means overtrained
Fin painting model probably not
MThx!
Shi, so i'm using automatic1111 on two computers and i'm trying to change the "commit" on one computer to match the other. how do i do that? i am not a programmer, i tried following a tutorial online but it did not work. am i even doing this correctly
Finside folder
Fuse git pull
Fthen
Fuse certain commit id
Fi explained in this video
FYouTubeSECourses - Software Engineering Courses
Discord : https://bit.ly/SECoursesDiscord. New fantastic style transfer feature via T2I-Adapter added to the #ControlNet extension. If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on 
https://www.patreon.com/SECourses
Playlist of #StableDiffusion Tutorials, #Automatic11...
https://www.patreon.com/SECoursesPlaylist of #StableDiffusion Tutorials, #Automatic11...
Fhow to use certain commit and git pull
Sah wow, ok that just answered everything for me. i was trying to use git bash to do it and was failing. thanks!
Mdo I need to pay money
JThanks for the welcome, specifically, would I be able to train a model based on my own video footage to generate video footage based on insights into the motion in the input footage?
PThanks very much for all the tutorials. They are amazing!
I am looking to know how to train a model based on a selection of images from a photographer to capture the same tones, light and style. What would be the recommendation?
I am looking to know how to train a model based on a selection of images from a photographer to capture the same tones, light and style. What would be the recommendation?
Fnot that advanced yet
Fthis video would be my recommendation to see best settings
FYouTubeSECourses - Software Engineering Courses
RunPod: https://bit.ly/RunPodIO. Discord : https://bit.ly/SECoursesDiscord. Web UI DreamBooth got epic update and we tested all new features to find the best settings (10). If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on https://www.patreon.com/SECourses
Playlist of...
https://www.patreon.com/SECoursesPlaylist of...
Fmoreover you can try to train with 1024x1024
Frunpod is paid service yes. for colab it has free option
Mok thx
PThanks Dr. Furkan, what would be your recommendation for classification images and the captions for the style?
MMhow can I fix this
Shi,i get black images when i try to generate images from lora trained model , i uninstalled xformers,and disabled it but no help
MI think the checkpoint your using has something to do with the problem
MI found the problem, turns out I didn't have enough storage
H@Furkan Gözükara SECourses when you have to move to new pod is it best to just transfer entire SD folder using Runpodctl?
Hor folder by folder
Fentire folder easier
Fi use the class token to generate classification images with cfg 7
Ffile captions i think necessary when you want to improve quality of the model. for faces not needed. but if you want to teach full body, perhaps clothes and backgrounds can be defined if they are not different in every image
Fwhere did you train the model
HJust tried, takes a very long time
HWget for models is super quick
