if you use other images than the model you are training on, it will do also fine tuning according to
if you use other images than the model you are training on, it will do also fine tuning according to those new images
Fso the model original knowledge of that class will change
RSo one should try to use images generated by the base model used for the Lora in order not to mess up the models own fine tuning. I'm going to test this out
Acan someone help me with this error on runpod?
message.txt16.42KB
Fwelcome
Fwhat are you trying to do_?
Ai try to launch relauncher.py with python relauncher.py
Ai just finished setting up my runpod using your videohttps://www.youtube.com/watch?v=zA4LksIVas8&t=334s
YouTubeSECourses
RunPod: https://bit.ly/RunPodIO. Discord : https://bit.ly/SECoursesDiscord. How to install latest version of Automatic1111 Web UI and #DreamBooth on #RunPod. 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 #StableDiffusi...
https://www.patreon.com/SECoursesPlaylist of #StableDiffusi...
Abut before i never encountered this error
Ranother question. Batch size will affect how exactly the learning rate? I know one should divide total steps by batch size to get the number of steps. So for a training of (12 instance images x 10 epochs x 80 repeats) (total 9600steps) / batch size = 1200 steps... I use batch 8 to speed up the process, but does this degrade the learning rate or loss? And I guess 1200 steps is on the low side. I should be rather between 1500-3000 steps after batch divide, correct?
Awhat.... now it worked, i haven't done anything, just stopped and started my pod...?? maybe some bugs related to runpod??
JHey guys. Thank you Furkan for your contributions!
I've a question. I'm running Dreambooth trainings with the A111 SD extension. I've a Nvidia GeForce RTX 3060 12gb. For what I've seen from your videos, the card should be able to make 5-7 it/s right? I don't know if anyone here has the same card and does trainings...
I'm stuck in 2.5 it/s and can't go any faster, I've tried a lot of things. Do anyone here know what things I should test or check to see if there's any issue or that's the expected card speed?
I've a question. I'm running Dreambooth trainings with the A111 SD extension. I've a Nvidia GeForce RTX 3060 12gb. For what I've seen from your videos, the card should be able to make 5-7 it/s right? I don't know if anyone here has the same card and does trainings...
I'm stuck in 2.5 it/s and can't go any faster, I've tried a lot of things. Do anyone here know what things I should test or check to see if there's any issue or that's the expected card speed?
Fbatch size makes things different. it changes how the model learns : https://www.baeldung.com/cs/epoch-vs-batch-vs-mini-batch
Baeldung on Computer Science
Explore the differences between an epoch, a batch, and a mini-batch.
Fi think for teaching a face batch size 1 works best
Fit is for generating images not training
Ftraining slower
RThanks. will read up on this
Fso instead of showing a picture at a time
Fchanging weights
Fyou show multiple images to model
Fchange weight according to all
DHi Furkan, saw your Lora dreambooth video and wondered if you could share the CKPT file of your trained facf
D*face
DI'd like to play around and see how photorealistic it is myself.
DRefering to this video https://youtu.be/TpuDOsuKIBo
YouTubeSECourses
#Kohya SS web GUI DreamBooth #LoRA training full tutorial. You don't need technical knowledge to follow this tutorial. In this tutorial I have explained how to generate professional photo studio quality portrait / self images for free with Stable Diffusion training.
Our Discord server
https://bit.ly/SECoursesDiscord
If I have been of assis...
Our Discord server
https://bit.ly/SECoursesDiscord
If I have been of assis...
FMy own face ckpt?
FWhy would you need that
FBy the way I did training realistic vision model
FDownload link already in gist file
JHey! Love the tutorials. Im getting stuck at generating images in Automatic1111 getting "OutOfMemoryError: CUDA out of memory." tried a bunch of things but havent been able to solve it yet. Any ideas?
Tmeans you dont have enough vram
Jis the only real solution to use a better graphics card?
Jor is there some settings i can use that help reduce the load on VRAM
Tget a bigger/better card or use runpod/google collab. there are many techniques @Furkan Gözükara SECourses shows in his video many techniques to lower vram make sure to watch closely
Jah got it, thanks!
TKoyha ss from this video. https://www.youtube.com/watch?v=TpuDOsuKIBo&loop=0
Getting this error, I have tried many things using GPT 4 and cant get it to work.
I am using cuda 11.8, proper cudnn files. Card is a gtx 1070, fresh nvidia install.
Getting this error, I have tried many things using GPT 4 and cant get it to work.
I am using cuda 11.8, proper cudnn files. Card is a gtx 1070, fresh nvidia install.
Tthe installations succeeds, but training fails with this error
HCould somebody help me with my vladmatic install i was having issues with diffusion storing to much vram so i deleted the venv to reset it but now im getting an issue with it not recognizing clip?
RYour resolution setting might be too high. Set it at 512x512 and try again
Dis it possible to batch sd upscale a folder of images but with different prompts ? as if you were to load the image in to png info , send to img2img and sdupscale using prompt from png info , but in a batch process?
Fyes you can reduce vram usage greatly
Ffirst try
FFif still not works add
Fdon't forget to use too
Fthee are all command line arguments
Fwhat options have you picked? exactly as me?
Fno idea about this if there is such extension or not :/
Fi am working on high quality sorting algorithm of generated images tutorial
Fso lets say you generated 3000 images of your self