Best Settings For The END of Photography - Use AI to Make Your Own Studio Photos, FREE Via DreamBooth Training
Added 2023-07-04 17:41:18 +0000 UTCTutorial: The END of Photography - Use AI to Make Your Own Studio Photos, FREE Via DreamBooth Training
Join discord and tell me your discord username to get a special rank : SECourses Discord
Github readme file : click to open
Best Settings
Click screenshot links to see them
Or download attached zip file which contains all images
SS_1 , SS_2 , SS_3 , SS_4 , SS_5 , SS_6 , SS_7 , SS_8 , SS_9
Comments
not yet but i will hopefully make research for this. stay tuned. sorry for delay
Furkan Gözükara
2024-01-03 10:05:49 +0000 UTCDo you have a settings file that will work for Kohya for SD 1.5 similar to the one you provide for SDXL dreambooth training?
Khoa Vo
2024-01-03 07:53:30 +0000 UTCYes you can delete all
Furkan Gözükara
2023-12-24 11:36:21 +0000 UTCThank you for that! any paths were I can delete the checkpoints after the model is finished? I assume I don't need to keep all of them
Luis Arellano
2023-12-24 04:23:31 +0000 UTCif you delete with jupyter lab they will be sent to trash bin execute this : rm -rf /workspace/.Trash-0
Furkan Gözükara
2023-12-22 16:06:54 +0000 UTCI've been deleting files but when I train the models, these end up taking so much space. is that normal? how much space do you usually need to train these models? i'm using 12 photos and the settings you described. Any other folders that I should delete?
Luis Arellano
2023-12-22 15:54:07 +0000 UTCyou can delete folders and objects with rm -r file_path you can also turn off machine and expand your storage
Furkan Gözükara
2023-12-22 14:23:12 +0000 UTCIt's weird, I did selected a different (i assume checkpoint) from the trainings and that worked. Now however, i get a different error. No space left on the device. I'm using podrun. 40 GB Disk / 75 GB Pod Volume. I though that would be enough but seems that training these models takes up so much space.. any good advice for how to clear space? should I delete some of the checkpoints from my training?
Luis Arellano
2023-12-22 13:08:42 +0000 UTCthat is very weird. where you are trying to generate on automatic1111? you can use your trained model on the latest version of automatic1111. so have 2 installations. 1 latest 1 dreambooth extension
Furkan Gözükara
2023-12-22 10:50:39 +0000 UTCI just did a trainning, but when trying to generate an image I get this error: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! any ideas?
Luis Arellano
2023-12-22 08:46:47 +0000 UTCit increases the emphasis of the prompt explained here : https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#attentionemphasis
Furkan Gözükara
2023-09-25 23:54:24 +0000 UTCWhat's the role of the "1.1" in the inference prompt? Eg: (ohwx man:1.1)
Manpreet Singh
2023-09-25 23:41:06 +0000 UTCthat repo is not maintained anymore much. sadly i don't know
Furkan Gözükara
2023-09-25 09:25:18 +0000 UTCThanks!
Manpreet Singh
2023-09-25 03:43:02 +0000 UTCDo you have any recommendations on how to take these WebUI settings and use them directly on shivamshrirao's dreambooth repo (I think the DB extension is based on that) and run training and inference via a python script?
Manpreet Singh
2023-09-25 03:42:55 +0000 UTCyou are welcome. realistic vision 5.1 supposed to be better version of 2.0. sdxl is far superior than sd 1.5 if you do dreambooth training.
Furkan Gözükara
2023-09-25 00:14:25 +0000 UTCIs there a significant difference between 1) using Realistic Vision 2.0 and 5.1 and also 2) using SDXL instead of 1.5? Trying to see what to include in my personal "v1". Thanks for your amazing tutorials and detailed instructions btw!
Manpreet Singh
2023-09-25 00:01:39 +0000 UTCall same settings. i haven't found a better way yet. but i suggest you to do SDXL DreamBooth now instead of SD 1.5. hopefully a tutorial coming but config files shared here already : https://www.patreon.com/posts/very-best-for-of-89213064
Furkan Gözükara
2023-09-24 16:49:46 +0000 UTCOk, thanks for sharing. Do all other settings remain exactly the same or are there some differences?
Manpreet Singh
2023-09-24 16:13:00 +0000 UTCi suggest to use latest realistic vision which is 5.1 : https://huggingface.co/SG161222/Realistic_Vision_V5.1_noVAE
Furkan Gözükara
2023-09-24 12:13:36 +0000 UTCThe first setting screenshot SS_1.png shows Realistic Vision 3.0 with VAE as the model to use. Whereas the YouTube video talks about Realistic Vision 2.0. Which one would you recommend to use? And what are the differences between the two?
Manpreet Singh
2023-09-24 00:48:29 +0000 UTCyep leave it empty as in pictures
Furkan Gözükara
2023-08-14 10:41:36 +0000 UTCHi! what about "advanced" options on "settings"? I can leave it empty?
Tiago4D
2023-08-14 03:17:40 +0000 UTCyou are welcome. thanks a lot for the info
Furkan Gözükara
2023-08-01 08:39:58 +0000 UTCSo for everyone reading this, as Furkan pointed, reg (class) images need to be exactly the same resolution as the training images. I was using the 1024x1024 dataset instead of the 768x1024. Using the right one worked. On another note: Automatic1111 doesn't allow training Dreambooth on an inpainting model, and gives the error above, which is very verbose but it means basically that. If inpainting is required, then one can merge the resulting Dreambooth final model with the SD-1.5 inpainting model post-training. Thanks again Furkan
Daniel Limia Aspas
2023-08-01 00:34:57 +0000 UTCInteresting. Using a non inpainting version casues issues with the trained model when using it to inpaint the face of the subject. But maybe I can merge the sd-1.5-inpainting model afterwards
Daniel Limia Aspas
2023-08-01 00:14:27 +0000 UTCinpainting may not work. never tried it. i use regular version
Furkan Gözükara
2023-08-01 00:09:53 +0000 UTCThank you, I will apply your feedback. By the way, when loading Realistic Vision 4 (inpainting) as a source checkpoint and creating a new Dreambooth model, it fails to create the model. Does it happen to you too? Error: 2023-08-01T00:06:05.203719455Z size mismatch for conv_in.weight: copying a param with shape torch.Size([320, 9, 3, 3]) from checkpoint, the shape in current model is torch.Size([320, 4, 3, 3]).
Daniel Limia Aspas
2023-08-01 00:01:45 +0000 UTCYour issue is class images and train images not exactly same resolution
Furkan Gözükara
2023-07-31 23:40:08 +0000 UTCIf fp32 available prefer it. Check hugging face of that model. I would suggest you to test realistic vision 5.1. Just released recently
Furkan Gözükara
2023-07-31 23:39:41 +0000 UTCYes that is why. Resolutions have to be exactly same. Use 768 x 1024 I have them here https://www.patreon.com/posts/4k-2700-real-84053021
Furkan Gözükara
2023-07-31 23:38:52 +0000 UTCYep did a fresh install with the scripts I shared here : https://www.patreon.com/posts/runpod-auto-84716845
Furkan Gözükara
2023-07-31 23:37:10 +0000 UTCAlso, another question if you don't mind answering: is using fp32 base models necessary? or using fp16 is sufficient? Asking because currently one of the best realism models (epiCRealism) does not have a fp32 version
Daniel Limia Aspas
2023-07-31 23:36:49 +0000 UTCCan it be because i am training 768x1024 images? Maybe i should use the 768x1024 reg images dataset?
Daniel Limia Aspas
2023-07-31 23:35:58 +0000 UTCDisabling class matching seems to work? (not sure): https://github.com/d8ahazard/sd_dreambooth_extension/issues/1180
Daniel Limia Aspas
2023-07-31 23:33:53 +0000 UTCThank you! I am using the same dataset. Are you using the latest Stable Diffusion runpod template? (v10)
Daniel Limia Aspas
2023-07-31 23:33:09 +0000 UTCI just verified and it didn't generate any images. used 1024x1024 training images. used the images downloaded from here as class images : https://www.patreon.com/posts/4k-2700-real-84053021
Furkan Gözükara
2023-07-31 22:53:49 +0000 UTCwith 8 GB vram unfortunately you can only do LoRA training. i have a LoRA tutorial here : https://youtu.be/TpuDOsuKIBo
Furkan Gözükara
2023-07-31 22:17:30 +0000 UTChello. are your training images are also exactly 1024x1024? can you verify that? so sorry for late reply i was on vacation . i will test this myself now
Furkan Gözükara
2023-07-31 22:16:51 +0000 UTCIt turns out that if I have an 8 GB video card, then I will not be able to use this training? Or is it possible, but with different settings?
Alex May
2023-07-31 19:05:23 +0000 UTCHi! when using your classification images dataset posted, Dreambooth does not seem to pick them up well and it always starts generating new ones. I am using the 1024x1024 dataset, max resolution set to 1024px. I downloaded them to runpod using wget and they're in the right folder.
Daniel Limia Aspas
2023-07-31 12:57:34 +0000 UTCawesome ty so much. sorry for late reply i am on vacation. hopefully even better videos coming soon
Furkan Gözükara
2023-07-25 21:56:35 +0000 UTCawesome. yes cropper will just crop and leave max size. then you need to resize.
Furkan Gözükara
2023-07-25 21:56:05 +0000 UTCThanks, I think wget actually did solve it :) I had already used A1111 to resize as shown in the video. Love your content by the way. As someone learning this for fun, it's always very clear and easy to follow - thanks for all that you do!
Crypto Melon
2023-07-23 15:33:06 +0000 UTChow exactly did you upload them to the runpod? drag and drop may not work very well. runpodctl works great. moreover you should use wget to download into runpod. cropper.py will not resize them into the correct resolution. it will only crop to the target aspect ratio as much as possible with highest resolution. so please use automatic1111 to resize them into the exact resolution as i have shown in the video
Furkan Gözükara
2023-07-23 08:58:40 +0000 UTCHi, I have two problems (using runpod) 1. Dreambooth isn't finding my class images and is trying to generate new ones. My max res is "1024", class images are your "768x1024", and I used your cropper.py + SD training to make my training images "768x1024" also. I thought I got around this by turning on "Disable Class Matching" in the testing tab, but t hat created problem #2: 2. log error: 2023-07-23T02:58:08.048664108Z ValueError: 'Training Images/man_training/00812-0-eric-ardito-GdWwogJ4TOk-unsplash.png' is not in list. *I checked this file and it's able to be opened, everything seems fine with it. I've googled my way through other problems along the way, but I can't get past this. Any help would be massively appreciated! *edit: I just noticed that I didn't use your instructions to use the class images in runpod. I just drag and dropped them into a runpod folder that I created, after downloading and extracting on my pc. They do seem fine to me with this method, however.
Crypto Melon
2023-07-23 03:33:10 +0000 UTCYes, you're right. It's the problem of the resolution of the training images. But, I was processed these images by cropper.py, I thought it was correct but which is not. After I prepared my training images with birme, everything was OK.
Daniel Chen
2023-07-22 11:25:32 +0000 UTCyes Kohya can do both but implementation differs. DreamBooth extension fo Automatic1111 web UI can also train both DreamBooth and LoRA. so Kohya and that extension can be compared
Furkan Gözükara
2023-07-21 13:18:16 +0000 UTChello i am so sorry for late reply. the patreon didn't show me your reply. do you still need help? please private message me
Furkan Gözükara
2023-07-21 13:17:17 +0000 UTChello. sorry for late reply. this usually happens when the resolutions of images are mis matching. can you verify that resolution of class images, training images are all same? also you have set the max resolution of training?
Furkan Gözükara
2023-07-21 13:16:37 +0000 UTCI have generated about 2700 images by using t2i, and after I click the training button, cmd checked those images, and said we need about 2400 classification images and start generating … why? I think we are fine with the numbers
Daniel Chen
2023-07-21 12:16:09 +0000 UTCRAW photo, a portrait photo of (ohwx man:1.3), traveler clothes, standing in the forest, natural skin, 8k uhd, high quality, film grain, Fujifilm XT3 Negative prompt: (deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime), text, cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, BadDream, UnrealisticDream Steps: 25, Size: 640x960, Seed: 441072722, Sampler: DPM++ SDE Karras, CFG scale: 7
Максим Сутковенко
2023-07-15 15:03:46 +0000 UTCCan you write here prompts? I can test on my training and give you feedback
Furkan Gözükara
2023-07-15 10:55:24 +0000 UTCBut I trie to use realism prompts...
Максим Сутковенко
2023-07-15 07:41:17 +0000 UTCHello. I haven't tested thoroughly. But with realism not every prompt works. I am working on a styling workflow as well. Styling is different than realism.
Furkan Gözükara
2023-07-14 22:55:18 +0000 UTCHey, did you try another interesting prompts, which generate you? Because I can not any good results with other prompts, except for yours (provided in video.)
Максим Сутковенко
2023-07-14 21:09:01 +0000 UTCok probably that is why. if your train images are 512x512, make max resolution 512 and use 512x512 class images dataset
Furkan Gözükara
2023-07-12 14:47:45 +0000 UTCSure. Max res 1024. Image res train 512x512. 14 images training
Samuel
2023-07-12 14:40:18 +0000 UTCyour setup looks correct. may i learn your set max resolution? your training images resolution? also how many images you have? so sorry for late reply
Furkan Gözükara
2023-07-12 14:25:26 +0000 UTCWhy after putting : https://gyazo.com/07bea5d8a2b4831a354a3828ac18a31e And having: https://gyazo.com/c4871748b12e61de7797bf6352a56956 It's telling me I need to create 1400 images class and start generating??? I have the class images... I dont understand.
Samuel
2023-07-12 01:31:03 +0000 UTCSince Kohya can train both Dreambooth and LoRA, as well as anything else, really, what exactly are you going to compare?
Constantine Tenma
2023-07-08 04:50:01 +0000 UTCActually I find Kohya inferior than DreamBooth extension. I will hopefully make a new video for LoRA better settings as well and I plan to compare Kohya and also DreamBooth extension
Furkan Gözükara
2023-07-07 08:14:01 +0000 UTCWanted to thank you so much for this tutorial. I've been trying to train myself for ages, but as a long hair clean shaven guy the output would always end up too masculine using "man" class or feminine using "person". Using male reg images and the man classifier It looked great at 4200 steps. I know I didn't have has much luck with your LORA tutorial last time, but will try again. In that you're using 8bitadam instead of Lion like here. Do you have an updated configuration of training lora with Lion or will you update in the next few months? Also do you find Koyha is always best for LORA training?
Broadband
2023-07-07 03:23:24 +0000 UTC
