Software Engineering Courses (SECourses)•3y ago

Thoughts? Testing out new sdxl dreambooth settings, with a merge technique

Furkan Gözükara SECourses•12/5/23, 8:07 PM

testing 768

FFiveSAI Thoughts? Testing out new sdxl dreambooth settings, with a merge technique

Tesli23•12/5/23, 8:24 PM

link?

TTesli23 link?

FiveSAIOP•12/5/23, 8:24 PM

link? these are my own settings im testing

Tesli23•12/5/23, 8:25 PM

oh ok, how are you running dreambooth?

FiveSAIOP•12/5/23, 8:26 PM

Yep dreambooth, but suprisingly the settings somewhat work with lora too

FiveSAIOP•12/5/23, 8:28 PM

i've done about 600+ trainings past 2 months or so

Tesli23•12/5/23, 8:30 PM

iv'e done too many trainings as well. Are you running dreambooth in colab?

TTesli23 iv'e done too many trainings as well. Are you running dreambooth in colab?

FiveSAIOP•12/5/23, 8:31 PM

Nope on a 4090 on my local system

Tesli23•12/5/23, 8:31 PM

nice, i have a 4090 as well. we are the lucky ones

FiveSAIOP•12/5/23, 8:32 PM

Yep managed to get it before price surged like crazy

Furkan Gözükara SECourses•12/5/23, 9:36 PM

fully updated

Furkan Gözükara SECourses•12/5/23, 9:36 PM

https://www.patreon.com/posts/94098751

Furkan Gözükara SECourses•12/5/23, 9:51 PM

https://github.com/magic-research/magic-animate/issues/44

GitHub

Full Auto Installers and Instructions For Windows Including Raw Vid...

I have spent over 24 hours to prepare all these scripts You can download all from here : https://www.patreon.com/posts/94098751 Example 15 second DensePose I generated via my script 15_sec_512_clos...

740•12/5/23, 10:40 PM

missing nodes doesnt appers in manager

740•12/5/23, 10:40 PM

how can I install them

AIGambino•12/5/23, 10:46 PM

Instead of using the base sd xl for training with kohya could we use turbovisionxl for superior trainings?

Aldo Raine•12/5/23, 10:48 PM

@Dr. Furkan Gözükara I'm getting this error when Kaggle training now:
Your notebook tried to allocate more memory than is available. It has restarted.

Aldo Raine•12/5/23, 10:51 PM

that was after the first 20%/safetensor

DDR.Siva Someone help me with creating images with two different face lora

mtnmecca•12/5/23, 10:52 PM

focused on this today and so far this works. 1. Use Kohya to merge either two loras in ckpt or into one lora. 2. In adetailer use detection K of 2 with adetailer prompt similar to: person 1 <lora 1> [SEP] person 2 <lora2> 3. depending on the image generated with two people and which face has a stronger face detection then you may have to switch step 2 to get the proper detailing.

7740 missing nodes doesnt appers in manager

Furkan Gözükara SECourses•12/5/23, 11:43 PM

comfy ui manager

AAIGambino Instead of using the base sd xl for training with kohya could we use turbovision...

Furkan Gözükara SECourses•12/5/23, 11:43 PM

you can try and compare

AAldo Raine @Dr. Furkan Gözükara I'm getting this error when Kaggle training now: Your noteb...

Furkan Gözükara SECourses•12/5/23, 11:43 PM

you see 29 gb ram right?

FFurkan Gözükara SECourses you can try and compare

AIGambino•12/5/23, 11:58 PM

Have you tried this before? Using something other than base sd xl model

AAIGambino Have you tried this before? Using something other than base sd xl model

Furkan Gözükara SECourses•12/6/23, 12:02 AM

havent had chance yet

FFurkan Gözükara SECourses you see 29 gb ram right?

Aldo Raine•12/6/23, 12:23 AM

Yes 29GB max

Aldo Raine•12/6/23, 12:24 AM

only thing I did different this time was use stablediffusionapi/juggernaut-xl-v7

Aldo Raine•12/6/23, 12:24 AM

instead of sdxl base

AIGambino•12/6/23, 1:21 AM

Whats going on with this yolo script

AIGambino•12/6/23, 1:21 AM

AIGambino•12/6/23, 1:21 AM

Mmtnmecca focused on this today and so far this works. 1. Use Kohya to merge either two l...

DR.Siva•12/6/23, 2:03 AM

Will give it a try

Mmtnmecca focused on this today and so far this works. 1. Use Kohya to merge either two l...

DR.Siva•12/6/23, 4:01 AM

any special argument ? keywords ? etc to be used ? Or ohwx / bbuk woman and man keys will be taken from lora automatically ? I dont want to create a ckpt/safetensor , it will limit my usuage to be used with other good models. I will see if thre is possibility of merging them into one lora for two faces. Or @Dr. Furkan Gözükara is it possible to traing lora for two faces with captions ? That will be easy i guess ?

FFiveSAI Yep dreambooth, but suprisingly the settings somewhat work with lora too

Benjamin•12/6/23, 8:05 AM

sounds interesting, do you mind sharing your training settings?

FFurkan Gözükara SECourses comfy ui manager

740•12/6/23, 8:42 AM

They are not showing Up in comfyui manager

7740 missing nodes doesnt appers in manager

740•12/6/23, 8:42 AM

I have already installed all missing custom nodes from that tab, but still missing these

hozaggers•12/6/23, 10:24 AM

I am trying to follow the tutorial on https://www.patreon.com/posts/kohya-sdxl-lora-88397937
when I try to 'start training' I get an output
09:59:46-038102 INFO Start training Dreambooth...
09:59:46-039759 INFO Image folder does not exist

not clear to me which Image folder it doesn't like... here is my inputs and my kaggle file... what am I doing wrong?

Deleted User•12/6/23, 10:28 AM

Click on prepare training data firts then copy info to folder

hozaggers•12/6/23, 10:28 AM

aha did the trick thank you

Deleted User•12/6/23, 10:41 AM

You are welcome

hozaggers•12/6/23, 11:06 AM

does anyone know what might be causing the training to fail?

message.txt2.53KB

Bbaloooo My workflow: 1. I play in GUI 2. Spy with the GIT package above 3. Do iterative ...

andrsoe•12/6/23, 11:09 AM

wait so, you recreate your UI results with the json api call with 40% succesrate ? SUS

Hhozaggers does anyone know what might be causing the training to fail?

Deleted User•12/6/23, 11:28 AM

The number of images must take at least 30 to catche

Deleted User•12/6/23, 11:28 AM

Your catche must be finished inside 30 minutes

Deleted User•12/6/23, 11:29 AM

Your training killed in 1577 images catched so resuce your training dataset images to 1550

Hhozaggers does anyone know what might be causing the training to fail?

Deleted User•12/6/23, 11:30 AM

How many training inages in your data set

Elisoli•12/6/23, 11:30 AM

on magic animate i get this error and it stuck like this :

pipeline_animation.py:624: FutureWarning: Accessing config attribute in_channelsin_channels directly via 'UNet3DConditionModel' object attribute is deprecated. Please access 'in_channels' over 'UNet3DConditionModel's config object instead, e.g. 'unet.config.in_channels'.
num_channels_latents = self.unet.in_channels
0%| | 0/25 [00:00<?, ?it/s]

Deleted User•12/6/23, 11:33 AM

During extract lora from my checkpoint in kaggle i got error out of memory how to solve the problem and use gpu during the process

DDeleted User How many training inages in your data set

hozaggers•12/6/23, 11:39 AM

I have 174 images in training dataset, although the regularization I think has 5200? if I re-run will it continue to cache them until it caches them all? or does it start caching from the start each time?

Hhozaggers I have 174 images in training dataset, although the regularization I think has 5...

Deleted User•12/6/23, 11:40 AM

It will start catche from 0

DDeleted User It will start catche from 0

hozaggers•12/6/23, 11:42 AM

the error says 1577/5374 which I think makes sense. 5200 regularization images + 174 of my own training images. not anything to do with my internet upload speed since everything is already uploaded to Kaggle? it still seems quite a low number to timeout at (not even going through all the regularization images)

Thoughts? Testing out new sdxl dreambooth settings, with a merge technique

Similar Threads