Software Engineering Courses (SECourses)•3y ago

wget --quiet --show-progress https://github.com/runpod/runpodctl/releases/download/v1.9.0/runpodctl-

wget --quiet --show-progress https://github.com/runpod/runpodctl/releases/download/v1.9.0/runpodctl-darwin-arm -O runpodctl && chmod +x runpodctl && sudo mv runpodctl /usr/local/bin/runpodctl

justalurkerOP•3/8/23, 10:30 AM

@vjavad1443 github said that's how to install on a mac

Jjustalurker wget --quiet --show-progress https://github.com/runpod/runpodctl/releases/downlo...

vjavad1443•3/8/23, 10:30 AM

yes it didn't work

TTrimac Ey guys! In dreambooth do you have to select specific 1.5 model on top?

Furkan Gözükara SECourses•3/8/23, 10:31 AM

you can use any model but training results probably wont be good as using 1.5 because these models are baked

justalurkerOP•3/8/23, 10:31 AM

ok, https://github.com/runpod/runpodctl has various install directions, maybe something there will work

VVirus Do you guys know any information on how to utilize ControlNet to indicate the lo...

Furkan Gözükara SECourses•3/8/23, 10:33 AM

in this video i explained which part is which part you seen? https://youtu.be/iFRdrRyAQdQ

YouTubeSECourses - Software Engineering Courses

Fantastic New ControlNet OpenPose Editor Extension & Image Mixing -...

Our Discord : https://discord.gg/HbqgGaZVmr. In this video, I am explaining how to use newest extension OpenPose editor and how to mix images in ControlNet. If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on

https://www.patreon.com/SECourses

Playlist of #StableDiffusio...

TTrimac On going training on dreambooth. I wanted to cancel my training progress since i...

Furkan Gözükara SECourses•3/8/23, 10:34 AM

i think it saves last step in diffusers weights. if not it uses last checkpoint

PP1pSl1p https://www.youtube.com/watch?v=sRdtVanSRl4&t=662s

Trimac•3/8/23, 10:34 AM

Okay for 5hrs working on this video following this settings for 8bitAdaw with exact parameters. Well, it did not get me precise results on my subject images "face" it looks like the subject but not good at all. Therefore settings from other may not work for others. Im thinking it dependes on what gpu we are running. Im running rtx 3060 12gb latest drivers updated. The man used higher graphics.

Vvjavad1443 @Dr. Furkan Gözükara Hi Dr, I'm truly appreciate for your time and effort. I've ...

Furkan Gözükara SECourses•3/8/23, 10:35 AM

i say yes use xformers 0.0.17 and ema

Bbush i have a question. Anyone using dreambooth must've come across JOE PENNA dreambo...

Furkan Gözükara SECourses•3/8/23, 10:36 AM

i havent checked. but you can use only man or photo of man as classification. it works even without classification but i found classification improves styling capability for faces

Vvjavad1443 Does anybody know why I'm stuck in the loop when run "python relauncher.py" comm...

Furkan Gözükara SECourses•3/8/23, 10:37 AM

change the loop in relauncher.py

TTrimac Okay for 5hrs working on this video following this settings for 8bitAdaw with ex...

Furkan Gözükara SECourses•3/8/23, 10:38 AM

could be related to your training dataset

Furkan Gözükara SECourses•3/8/23, 10:38 AM

no backgrounds clothes should be same

TTrimac Okay for 5hrs working on this video following this settings for 8bitAdaw with ex...

Furkan Gözükara SECourses•3/8/23, 10:47 AM

my computer also has 12 gb rtx 3060. it really looks good. you should be able to get good results same as base test on that video

FFurkan Gözükara SECourses i say yes use xformers 0.0.17 and ema

vjavad1443•3/8/23, 10:51 AM

So in youtube video you said the address we should run those commands ( uninstall ) is located in /workspace/venv/lib/python3.10

but for me there is no venv in workspace folder

I can find it in this address:
/workspace/stable-diffusion-webui/venv/lib/python3.10

vjavad1443•3/8/23, 10:51 AM

Is that ok?

Vvjavad1443 So in youtube video you said the address we should run those commands ( uninstal...

Furkan Gözükara SECourses•3/8/23, 10:51 AM

in unix

Furkan Gözükara SECourses•3/8/23, 10:51 AM

it works wherever you run actually

Furkan Gözükara SECourses•3/8/23, 10:51 AM

i mean on runpod at least

Furkan Gözükara SECourses•3/8/23, 10:51 AM

so it is okay

Furkan Gözükara SECourses•3/8/23, 10:52 AM

i think it is because of the template

Furkan Gözükara SECourses•3/8/23, 10:52 AM

template links to venv so all command run on venv automatically

vjavad1443•3/8/23, 10:55 AM

yes, I'm working on runpod. also I see some changes. For example I selected SD v1.5 template & after I opened it I see there is two models :

SDv1-5.ckpt
SDv2-768.ckpt

Furkan Gözükara SECourses•3/8/23, 10:56 AM

they changed template

Furkan Gözükara SECourses•3/8/23, 10:56 AM

but still works very well

Furkan Gözükara SECourses•3/8/23, 10:56 AM

currently i am doing 3 experiment

Furkan Gözükara SECourses•3/8/23, 10:56 AM

on dreambooth

Furkan Gözükara SECourses•3/8/23, 10:56 AM

on that template

vjavad1443•3/8/23, 10:57 AM

and would you please answer to this question as well. your opinion is important to me

I wanted to know have you tried the below format for learning rate field instead of 0.00001 ?

0.05:10,0.02:20,0.01:60,0.005:200,0.002:500,0.001:3000,0.0005

Vvjavad1443 and would you please answer to this question as well. your opinion is important ...

Furkan Gözükara SECourses•3/8/23, 11:02 AM

i am using constant

Furkan Gözükara SECourses•3/8/23, 11:02 AM

that can train very quickly

vjavad1443•3/8/23, 11:06 AM

I got this error

Screen_Shot_2023-03-08_at_4.05.17_AM.png

vjavad1443•3/8/23, 11:07 AM

I run command the first for command without any issue

Screen_Shot_2023-03-08_at_4.06.42_AM.png

vjavad1443•3/8/23, 11:07 AM

I run the first four command without any issue *

FFurkan Gözükara SECourses i havent checked. but you can use only man or photo of man as classification. i...

bush•3/8/23, 11:08 AM

i was actually talking about the content of the images and not the name of class

FFurkan Gözükara SECourses but this machine consumes a lot credits

Arvy•3/8/23, 11:09 AM

I had my 40gb vram card run out of memory several times yesterday on Runpod. I switched to vast.ai and haven't faced an issue so far. However downloading from vast.ai is very slow.

bush•3/8/23, 11:10 AM

like here, they arwe all real photos from unsplash

vjavad1443•3/8/23, 11:11 AM

This command didn't work as well

Screen_Shot_2023-03-08_at_4.11.13_AM.png

Bbush i was actually talking about the content of the images and not the name of class

Furkan Gözükara SECourses•3/8/23, 11:15 AM

the images are generated by the name of class

AArvy I had my 40gb vram card run out of memory several times yesterday on Runpod. I s...

Furkan Gözükara SECourses•3/8/23, 11:15 AM

it totally depends on configuration

Bbush like here, they arwe all real photos from unsplash

Furkan Gözükara SECourses•3/8/23, 11:16 AM

if you want to fine tune model

Furkan Gözükara SECourses•3/8/23, 11:16 AM

improve quality

Furkan Gözükara SECourses•3/8/23, 11:16 AM

then you can feed better images

Furkan Gözükara SECourses•3/8/23, 11:16 AM

the classification image logic is

Furkan Gözükara SECourses•3/8/23, 11:16 AM

harder to explain quickly

Furkan Gözükara SECourses•3/8/23, 11:16 AM

but i am planning a video for that

Vvjavad1443 This command didn't work as well

vjavad1443•3/8/23, 11:17 AM

I used this command instead and it worked:

pip install xformers==0.0.17.dev464

Is it ok ?

Bbush like here, they arwe all real photos from unsplash

bush•3/8/23, 11:17 AM

gotcha. also, one more thing. so I have the deleted the pod on which i trained yesterday. but now I want to fine tune the model even more. the steps will be the same?

Vvjavad1443 I used this command instead and it worked: pip install xformers==0.0.17.dev464 ...

Furkan Gözükara SECourses•3/8/23, 11:18 AM

yep

Bbush gotcha. also, one more thing. so I have the deleted the pod on which i trained y...

Furkan Gözükara SECourses•3/8/23, 11:19 AM

you need to upload same model

wget --quiet --show-progress https://github.com/runpod/runpodctl/releases/download/v1.9.0/runpodctl-

Similar Threads

Similar Threads

Similar Threads