one question , how can we save the current state of setup ? in kaggle ?
one question , how can we save the current state of setup ? in kaggle ?
Fso many things to make tutorial of
Fyou save checkpoint and use it
Fnot state
DNo i meant , the notebook setup
Fyes on there
Fgive checkpoint file path
Flike
F/kaggle/working/mycheckpoint.safetensors
For in temp that one
Dtill kohya setup , it takes around 30-40 min , if something bad happens , the session is terminated or jupyter is off , all setup is lost
Di meant " !bash gui.sh --headless" till this line in code
Disnt this when we start using hte gui of kohya
Dis there a way to save till this state , the notebook ?
Dso that we dont want to wait everytime 30-40 min to install from first ?
Dalso can i turn on gpu before training ?
Dor is it needed while installing hte requiremnts to properly config ?
Dstarted the first cell now
Ddont want ot screenshot your notebook codes
Dfor first step itself it will take 15 min i guess.. i thought , if at all we can save the state of notebook at a specific point , it will easy for us to open it again and do necessary changes and start from where we left
Dended
Droughly 12 min
DDsolid 15 min till we reach gui
Dagain caching latent at 31:35 min , this too will fail i think , may be i will reduce few images from training ?
Donly 5 min to finish caching 
D83%|████████████████████████████████▌ | 1516/1818 [30:08<06:00, 1.19s/it]
[E ProcessGroupNCCL.cpp:828] [Rank 1] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=3, OpType=ALLREDUCE, Timeout(ms)=1800000) ran for 1802663 milliseconds before timing out.
[E ProcessGroupNCCL.cpp:828] [Rank 1] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=3, OpType=ALLREDUCE, Timeout(ms)=1800000) ran for 1802663 milliseconds before timing out.
D
Dreduced to 13 images - 521 steps , lets see
D23 min now ..
Fwhen this happens
Fis your notebook killed?
For only gradio
Fpixart-alpha (PIXART-α) looking amazing
prompt : a mighty lion running in new york city
styles
default, anime, digital art
hopefully full tutorial and auto…
prompt : a mighty lion running in new york city
styles
default, anime, digital art
hopefully full tutorial and auto…
CCan anyone help please, ive created about 10 'deep fake videos so far with no issues at all.
But the past few times my pc just crashes out and restarts about 80% sometimes 64% etc into processing.
But the past few times my pc just crashes out and restarts about 80% sometimes 64% etc into processing.
S@Dr. Furkan Gözükara I’m doing this as a hobby but there is ways to make money generating this images and continue the research? and cover the server cost? this is a nice discussion that I guess a lot of people will be interest
CI cant get a screenshot of my pc because I never know when its going to crash and with it taking a long time for each video I can't sit and watch it constantly 
ive screenshotted the YouTube tutorial.
ive screenshotted the YouTube tutorial.
CSo I select my face, choose my target video, check many faces, and everything runs smoothly then as I say it gets to 60/80 sometimes 90% then crash computer turns off? 
Fyour PC specs is the problem
Fcould be undervolted RAM
Fsadly i dont know. but some people making money
CI'm confused as it worked fine, till now
C12th Gen I5
32GB Ram
GeFroce GTX 1660ti
this is a screenshot of a previous video that worked perfectly. Im face swapping on the same video each time
32GB Ram
GeFroce GTX 1660ti
this is a screenshot of a previous video that worked perfectly. Im face swapping on the same video each time
Fwell try to increase ram voltage
Fa little bit
DIs close images datasets is good for dreambooth training or full body or half body images data set is good for woman
Fi prefer to add all
Fand after use adetailer during inference
FSthis is like a new stable difussion?
Fyes
