your 24gb config disables xformers... isnt that super slow?
your 24gb config disables xformers... isnt that super slow?
ABut will this know to center my person in the photo? It wont crop their head off?
MFurkan's script for Yolo does not cut the heads off, but cuts the people out of the picture first, so they are in focus. And the cropping script cuts out the images absolutely fine, so you don't have to worry about that, they help a lot in creating the perfect training set.
ADoes he discuss his scripts in a video by chance?
MMYouTubeSECourses
One of the most important aspect of Stable Diffusion training is the preparation of training images. In this tutorial video I will show you how to fully automatically preprocess training images with perfect zoom, crop and resize. These scripts will hugely improve your training success and accuracy.
Scripts Download Link
https://www.patreon...
Scripts Download Link
https://www.patreon...
MIf I understand you OK, it sounds like you want to switch the model used during Inpainting from the first model that generated the base image to a different, specialized model generated from Dreambooth.
If that's the case, you should be able to switch out models during Inpainting to use the other model OK.
If that's the case, you should be able to switch out models during Inpainting to use the other model OK.
S@Dr. Furkan Gözükara for the pix art the runpod script works now?
GAnyone know how to get the most recent version of dreambooth extention working on runpod with A1111?
Gnevermind I figured it out
Fit looks lower quality
Fhaha
FSDXL Turbo all you need to know about quality
By the way SDXL 20 steps vs 50 steps has 0 difference in terms of quality in like 99% of cases
By the way SDXL 20 steps vs 50 steps has 0 difference in terms of quality in like 99% of cases
Fsadly i dont know. i havent seen anywhere in documentations
Fyes slower but better quality
Fyes working
Fi recorded video editing
Fimproved gui as well
ADoes anyone know why I'm getting tons of weird malformed limbs?
AI'm using these negative prompts
bad anatomy, bad hands, three hands, three legs, bad arms, missing legs, missing arms, poorly drawn face, bad face, fused face, cloned face, worst face, three crus, extra crus, fused crus, worst feet, three feet, fused feet, fused thigh, three thigh, fused thigh, extra thigh, worst thigh, missing fingers, extra fingers, ugly fingers, long fingers, horn, extra eyes, huge eyes, 2girl, amputation, disconnected limbs, cartoon, cg, 3d, unreal, animate
bad anatomy, bad hands, three hands, three legs, bad arms, missing legs, missing arms, poorly drawn face, bad face, fused face, cloned face, worst face, three crus, extra crus, fused crus, worst feet, three feet, fused feet, fused thigh, three thigh, fused thigh, extra thigh, worst thigh, missing fingers, extra fingers, ugly fingers, long fingers, horn, extra eyes, huge eyes, 2girl, amputation, disconnected limbs, cartoon, cg, 3d, unreal, animate
Fsdxl 1024 x1024?
AYep
A
Fnope
Fyou are doing 2048x2048
Fmake it like 1.25 x max high res fix
Fthis is a known issue
Fmodels cant generate much higher resolution than they were trained
Fi got to sleep hopefully see you tomorrow
Aty for the help! sleep well talk tomorrow
Ehorrifying 
DIs animatediff the best option for animation or is there other options? Just wanted yo before I start
DAlso can we use sdxl trained Lora for animation which I think is using sd1. 5
MI have 2 body fix negative prompts.
First this:
"deformed body, bad hands, ugly hands, mutated hands, deformed hands, missing hands, too many hands, too long hands, bad legs, ugly legs, mutated legs, deformed legs, missing legs, too many legs, too long legs, bad fingers, mutated fingers, deformed fingers, missing fingers, ugly fingers, less than 5 fingers, too many fingers, too long fingers"
and a longer:
"more than 2 hand, more than 2 legs, less than 5 fingers, more than 2 breasts, more than 2 nipples, more than 2 knees, leg lock, bad anatomy, deformed body, ugly body, too long body, blurred body parts, smudged body, too long neck, cropped hands, cut hand, bad hands, ugly hands, mutated hands, deformed hands, smudged hands, missing hands, too many hands, missing hand head, cropped legs, bad legs, ugly legs, mutated legs, deformed legs, smudged legs, missing legs, too many legs, bad fingers, mutated fingers, deformed fingers, missing fingers, ugly fingers, too many fingers, bad thighs, faulty thighs, ugly thighs, mutated thighs, too wide thigh"
First this:
"deformed body, bad hands, ugly hands, mutated hands, deformed hands, missing hands, too many hands, too long hands, bad legs, ugly legs, mutated legs, deformed legs, missing legs, too many legs, too long legs, bad fingers, mutated fingers, deformed fingers, missing fingers, ugly fingers, less than 5 fingers, too many fingers, too long fingers"
and a longer:
"more than 2 hand, more than 2 legs, less than 5 fingers, more than 2 breasts, more than 2 nipples, more than 2 knees, leg lock, bad anatomy, deformed body, ugly body, too long body, blurred body parts, smudged body, too long neck, cropped hands, cut hand, bad hands, ugly hands, mutated hands, deformed hands, smudged hands, missing hands, too many hands, missing hand head, cropped legs, bad legs, ugly legs, mutated legs, deformed legs, smudged legs, missing legs, too many legs, bad fingers, mutated fingers, deformed fingers, missing fingers, ugly fingers, too many fingers, bad thighs, faulty thighs, ugly thighs, mutated thighs, too wide thigh"
MParts about the face are deliberately not included, because they may distort the Lora of the person
MYep, it's 512x512 model. 
SNo... I just want to further train inpainting model in UI
Ehas anyone tried this before/know how they are able to train so insanely fast?? https://dreamlook.ai/dreambooth
Ei did a sample test and its 75% faster than a 24gb machine on runpod
AGood morning, I have a question about training multiple characters in a single LORA.
A few weeks ago, you mentioned that when captions are added to images (and these include the concept, for example, ohwx), the class token from the name of the directory where the images are stored is no longer used. That is, the 'ohwx' part of '20_ohwx' is not used, only the '20' is used to determine repetitions.
Based on this, if I want to train multiple characters, which strategy is better?
a) One folder for each character, even if they all have captions.
b) A single folder that encompasses all characters.
On another note, let's assume that for character1 I have 50 images and for character2 I have 100 images. Should this be taken into account in some way? For example, adjusting the repetitions so that the multiplication of repetition and number of images is similar across all datasets, or would this be counterproductive, and could it cause the dataset with fewer images and more repetitions to be overfitted compared to the rest?
Best regards and sorry for the long message
A few weeks ago, you mentioned that when captions are added to images (and these include the concept, for example, ohwx), the class token from the name of the directory where the images are stored is no longer used. That is, the 'ohwx' part of '20_ohwx' is not used, only the '20' is used to determine repetitions.
Based on this, if I want to train multiple characters, which strategy is better?
a) One folder for each character, even if they all have captions.
b) A single folder that encompasses all characters.
On another note, let's assume that for character1 I have 50 images and for character2 I have 100 images. Should this be taken into account in some way? For example, adjusting the repetitions so that the multiplication of repetition and number of images is similar across all datasets, or would this be counterproductive, and could it cause the dataset with fewer images and more repetitions to be overfitted compared to the rest?
Best regards and sorry for the long message
VFurkan, here? /opt/conda/lib/python3.10/site-packages/scipy/init.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.24.3
warnings.warn(f"A NumPy version >={np_minversion} and <{np_maxversion}"
/opt/conda/lib/python3.10/site-packages/scipy/init.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.24.3
warnings.warn(f"A NumPy version >={np_minversion} and <{np_maxversion}"
usage: sdxl_train.py [-h] [--v2] [--v_parameterization]
warnings.warn(f"A NumPy version >={np_minversion} and <{np_maxversion}"
/opt/conda/lib/python3.10/site-packages/scipy/init.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.24.3
warnings.warn(f"A NumPy version >={np_minversion} and <{np_maxversion}"
usage: sdxl_train.py [-h] [--v2] [--v_parameterization]
Valways failed in checkpoint training
MThis is what a strong but flexible Lora graph looks like on Tensorboard. If a spike jumps out into the 3's that's borderline, but no more than 3's. So the goal is not to get max_norm/keys_scaled above 2. Plus a max_norm/max_key_norm should reach 1 before half of the training session, so you have time to normalize the model. So you can tell by looking at these values during training if the Lora will be good or not.
MI trained this model with d_coef=0.85, because the first version (with 1.02) was strong, it went to 3 several times.
MAnd the first tests showed that even this version was a bit strong.
Fit is not faster
Fthey use bigger batch size and lesser steps
Factually when I tested before they did training on 512x512 haha
Fthe second question is the answer. you want each dataset to be trained total number of steps equally
Fthat is the logic of repeating
Fotherwise if that is a balanced dataset put all into a single folder