cant tell need research
cant tell need research
ZDoc you have a recommended way of downgrading cuda from 12 to 11?
Fif nothing broken keep it
ZWell trying to run pinokio browser and stuff never installs right. Thinking it may be a cuda issue
Fye i never used it
Fyou can uninstall cuda from program add removes
Fdownload cuda 11.8
Finstall it
Fit becomes defualt
Zi'm on linux so yeah that won't work. I'm a try your runpod script for installing vid2desnepose
Fmy runpod scripts
Fwill work without any issues
Fjust change folder paths
Fi got to go see you later hopefully
Zhave a good one doc and thanks
AAny idea what this error is
T@Dr. Furkan Gözükara hjx, can u help me
Phi
PAnyone having trouble using the provided class files? my Dreambooth tries to generate the class even though it already exists in the directory. I think its occouring only class 'man' for a 'girl' its ok.
DNeed 512*512 Lora training setting for animatediff.
We trained our faces for sdxl and dreambooth..
We trained our faces for sdxl and dreambooth..
His there any advice for training in kaggle when your input images are a variety of dimensions? A lot of my training images are larger than 512512 but smaller than 10241024, and not necesarily square. Resizing them can give them unrealistic/stretched proportions, not sure if that might impact the training? or is it not possible to train these?
DThere is image resize script on the patreon
Hbut question is will resizing help if the proportions are all distorted?
Hlike taking a 600x800 picture and resizing to 1024x1024 will require different stretching proportions and it can make a full body shot look strange when you force it into that size
TYou can also just use birme.net. Way easier than command line
Finstall error. i would a fresh install. you can also manually install bitsandbytes
Fhi what help you need?
Fah with comfy
Fsadly i cant help
Fi have standalone installer that works
Fyes happens when mismatch of resolution
Feither training images or reg images res mismatching
Fmakes sense. i should research it
Fyou can try both of these
Fenable bucketing
Fwithout enable bucketing
Fyour only chance is first upscale and then crop. or as i said use bucketing and give it a try
DWish there was way to convert /downgrade lora.. But I think it will only take half time to train Lora. With 512 trained Lora, we can use with animatediff. Also by training it, we could have trained for all three models
Tso sad. thanks
FLoRA training is not that fast
Fit is like 1.3 second / it vs 1.6 second / it
Fye comfy is hard
HI'm thinking upscale and crop may still pose a problem if the aspect ratio is not square on the original image. I'm thinking I may need to sacrifice some detail/body on those in order to get a square fit? Does your resizer script help with that or will it not cut out body parts to make it square?
Fi have 2 scripts
Fwhen you run both
Fyou will get square
Fwith best quality
Hthanks, will I need to upscale first to above 1024x1024 if the images are smaller than that size before running the scripts?
MI would also like to say a few words about the sample photos. I am currently training a Lora, where all the training photos are upper body shots. My training burned very fast and I just went down to a d_coef of 0.65 to get a good lora. So that's why it's good to take your sample shots from as many angles and distances as possible, because you can make a stronger and more flexible Lora. Just taking the same type of photos will quickly get stuck and the model will learn too much.
ZGitHub
Mamba-Chat: A chat LLM based on the state-space model architecture 
- GitHub - havenhq/mamba-chat: Mamba-Chat: A chat LLM based on the state-space model architecture
Ait works fine with sdxl but not regular sd1.5 training, weird
