i trained with about 20 for the model i made of my self but the poses and angles didn't seem nearly
i trained with about 20 for the model i made of my self but the poses and angles didn't seem nearly as diverse
SI try to follow the new tutorial but I run out or memory error
Syeah needs to be different
Sya i just don't have many professional photos of myself hanging around
Sat this point you don’t need that those models are taking care off
Sya but selfies are all mirror image not real image.
Sthat’s right
Smy working theory atm is that that higher quality images and resolution get better results
Si tried a few celebs to experiment and i didn't like those nearly as much
Sdo you have you own card? or you are using runpod or kaagle
Severything local
Si figured buying a new pc would be cheaper in the long run
Si need to try and experiment on product photography again but i didn't have much luck with products that had text on them in the past
ELooks really good, how many repeats did you do for 220 images?
Eor was it 1 repeat and the rest just epochs?
Si just did 150steps (epochs) not sure what repeat is
Eso no reguleraztion images at all, 220 images * 150 33000 ste[s. did you not get overtraining issues? or did you choose an earlier checkpoint?
NHello, long time ago I made models and already forgot how to customize them in colab,
I used to make initial descriptions in WD14 loras and edit them further. How can I do the same in colab?
If you can, can you share the config for better training accuracy
I want to put the training 768 x768 pixels, base model take anylora from lykon
I used to make initial descriptions in WD14 loras and edit them further. How can I do the same in colab?
If you can, can you share the config for better training accuracy
I want to put the training 768 x768 pixels, base model take anylora from lykon
Sno i didn't like the early checkpoints, used the 33k one
Sonce it's finished i usually merge it with another model i like at about 30% to help clean it up a bit
Every nice workflow 
looks really really good
looks really really goodS@Dr. Furkan Gözükara anyone know if gradio link is working again yet?
Nsdxl gives very good results training specific objects, very versatile
Snice
S
Strying to make myself some wallpapers
SMy training Starts! I jus change the server to 1 x RTX 4090 8 vCPU 31 GB RAM!
S3hrs to go!
SThis is really nice!
R@Dr. Furkan Gözükara have you tested xformers/gradient checkpointing as well as Memory Efficient Attention? I am running without Gradient CH. and xformers but still have memory efficient attention on. Not sure if it decreases quality
RAlso, wit A6000, on runpod, running 14 images dreambooth LoRA, at 4 batch size, I run into OOM, without gradient checkpointing and xformers. Using class images. Any thoughts if there are any memory optimizations better than these two? What do you recommend @Dr. Furkan Gözükara ??
F220 pro photos is the key
Fi dont knoıw if colab supporting captions never tried
Fbut colab is working
Fi did training few days ago
FYouTubeSECourses
Our Discord : https://discord.gg/HbqgGaZVmr. How to do free Stable Diffusion DreamBooth training on Google Colab for free. 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 Stable Diffusion Tutorials, Automatic1111 and Goo...
https://www.patreon.com/SECoursesPlaylist of Stable Diffusion Tutorials, Automatic1111 and Goo...
Fi havent tested but it wasnt working like yesterday
Fit is super strong
Frtx 4090 sper strong on unix
Fi think memory efficient attention doing nothing 
Fdont enable gradient checkpointing
Fenable xformers
Ftry to get max batch size
F= max speed
Rbs 4 i did gradient + xformers. Speed is great already, around 3h compared to 6 hours for batch 2. I will try bs4 with xformers only.
Rquality is pretty awesome from epoch 1 already
Rby the way, what is your understanding of max steps. If batch 1 is good at 8000 steps, when you go to batch 2, you have a total of 4000 steps. Do you have a formula already for how many extra steps to give the higher batch to compensate for loss in quality? Generalisation though seems to improve up to a certain number of batch increases.
Rok, xformers only, no gradient checkpointing goes OOM
Sreally impressed by how cinematic some of these are coming out
Fthere is no formula