oh sorry I thought you were talking about a different dataset. I'm going to use this dataset and try
oh sorry I thought you were talking about a different dataset. I'm going to use this dataset and try it out using massed compute now.
Uupdate -Dwayne johnson training works fine. I get good results around the 3000th step. I will try again for other characters to see if matches with the training data or not. Thanks for your help. I have one more question, is there a way to programatically select the best checkpoint instead of manually checking the checkpoints to see which one is the best?
UAnother question - In the dataset Dwayne johnson does not have hair but when I generate images a few images for Dwayne Johnson hair is being added, do you know why and how can we keep it real? I can add no hair in the negative prompt but is there a different way to keep it generic without hardcoding.
EHI! I have a question about lora-training and captioning. A friend of mine suggest that when you caption your object your not supposed to mention the object in the image at all, only everything around in like a semantic mask and where the object-tag works as the caption for the lora-object. On youtube-tutorials on the other hand everyone I've seen says that you should in detail concistently describe the object and only mention shortly the rest of the image.
What would you guys say is the case here?
What would you guys say is the case here?
Fyou need to evaluate yourself
Fyes it is about prompt
Fjust add bald :d
Fcaptioning is needed only when you do a general fine tuning
Fohwx man , ohwx class works best
Fthen after training in details give your prompt
Fdetailed captioning only reduces your trained subject resemblance
Fbecause even if you caption the elements you dont want to be learnt, it still leaks info of what you want to be learnt
EOk, so no captions only caption tag in kohya? Should I use a high or low network dim for 50 images that are very consistent, I'm training a bag
Fi recommend use my config and dont change
Fyou can use folder name as caption
Fso like
F1_ohwx man
For have imagefile.txt files and write each one of them like ohwx man
Fboth works
EEx 24GB_GPU_Quality_Tier_1_22950MB_14.1_Second_IT_Trains_T5_and_T5_Attention.json ?
Fthat is config
Ffolder name means
F
Jdoes this matter? the outdated version
Fnope
Ffolder name doesnt matter we use latest version
UThank you 
The finetuned checkpoint is about 23gb, is there a way to reduce the size without losing the quality?
The finetuned checkpoint is about 23gb, is there a way to reduce the size without losing the quality?Fyou can do lora
Fi dont remember if supporting save as fp8
UI tried lora but I like the outputs that im getting from finetuning. I see fewer overfits and more consistency in finetuned model.
Fye fine tune best
EI'm sonfused and have to ask again,
am I wrong in that you still can't train a full checkpoint below 80 gb vram? And finetuning means full checkpoint right?
am I wrong in that you still can't train a full checkpoint below 80 gb vram? And finetuning means full checkpoint right?
EThe lora turned out really bad. No shape from prodct at all, it only makes like a genral bag like it dosn't under geometry, or the shape is to complex
Ffull checkpoint can be trained as low as 6 gb vram
F6GB_GPU_5400MB_14.1_second_it_Tier_3_512px.json
Fthis is same as 48 gb config quality
F8GB_GPU_7500MB_20.5_second_it_Tier_1.json
ECould i use the 24 gb config for my rtx 4090?
F100%
Fpay attention to your empty vram
Fit uses 23 gb
Fyou can increase block swap count if you cant have 23 gb free due to other apps
EOk, could I do say 6 different colors of the same bag (same shape) in one checkpoint? I guess I would need captioning then.
Fi think you can train different colors of same bag
UHello,
I'm trying to run grid generator but I get this error, do you know why?
I'm familiar with comfyui but this is my first time trying swarmui
I'm trying to run grid generator but I get this error, do you know why?
I'm familiar with comfyui but this is my first time trying swarmui
U
Fhard to know without seeing all
Fyou can post debug logs here
UI was able to get it working. I pulled the latest code and it started working.