how many epochs is good for small dataset of face photography
how many epochs is good for small dataset of face photography
Zquick question, once learning rate shows about 0.2, should we stop training?
Fthat looks like a training dataset error
Fcan you try with our example training dataset. this doesnt look correct at all
Fyes much better
Fawesome quality
Fthis totally depends on training dataset and learning rate. so you should compare checkpoints
Fnow i need to come up with a good thumbnail for gpt4 video
UWhat exactly do you mean by that? I have also tried it with a random test-dataset that i downloaded. Fileformat etc shouldnt be an issue right?
Fit can be. i saw 1 time it was not able to read training dataset files
Fi dont know maybe something in text file maybe image file
UAllright thanks, ill try some other things. Also i am working on MacOs, which isnt really optimal.
Zi wasn't using 0.005, i was using 0.05 learning rate which is probably why, changed it to 0.005 and now they look much better, should i stop before it gets to 0.2 learning rate
Z1500 steps at a 0.005 learning rate loss Strength 0.1404, wondering if i need to lower my learning rate cause it's creeping up and not sure it's trained well enough yet
Zthis is her original picture, it's getting close
Zright now i'm at 2000 steps and my loss rate is about 0.1606, is that too high? should I lower my learning rate, or just stop the training and go from here?
Zi'm assuming it's a sort of balancing act to get it perfect
Fah mac os. i wonder if that is the problem
Fyes lower learning rate can be even better
IHello everyone, I'm new here.
IThnk you SEOCourses, you made the best, most extensive, and ONLY valid tutorials for explaining A1111, Runpod, training etc.
IDr. Furkan, are you the man behind those tutorials?
Nwhere to start, what is your course suitable for this?
II'm still facing the last problem on Runpod, but I can honestly say that these tutorials are the best, and the only ones that go into details STEP-BY-STEP and you learn so much, if you have the patience to watch the tutorials, pause, and exercise.
II hope I can get this Runpod error fixed today, though.
ESomeone have any idea what prompts I can use to have my subject be generated whole on one picture? My training pictures are fine and presents whole object. But when I'm trying to generate it is just zoom in or in a different perspective
IIn my experience "full body shot" or "full body portrait" but you also need to change ratio, like 512x768. In general it is not easy for me either, most images are closeups.
Fhello welcome
Fyes i am
Fstart collecting training images
Fhopefully my style teaching video will be published this week
Fhi can you show the runpod error you are getting
IFirst of all thank you for all the great work, but especially for being available to give support to your viewers.
IMany others don't even bother replying to comments on YouTube, so we appreciate that.
IBut just 5 minutes ago I think I found what I did wrong. I was updating requirements file, not requirements versions.
II will do another session again, and let you know...
Zif i do lower learning rate for longer, will i get better details on my training
Fbut it will take longer time
Zlike i got really good results, but she's like, "The eyes are a little off" and what not
Zit's really close tho
Zmay train with 0.0005 next time or somethin
IFor the sake of reporting, this was the error I was getting. Exception initializing accelerator: Accelerator.init() got an unexpected keyword argument 'project_dir'
IMaybe it can be useful. In my case after many hours online, I got easily distracted and didn't see that there are two files: requirements & requirements versions
IIf the training starts and finishes correctly, I will share my experience with you Dr Furkan.
IFirst update: In Dreambooth optimizer 8bit Adam is not there anymore. I have 1. Torch AdamW 2. Lion
Fyes this requires experimentation
Fyep this is version error
Fi spent like 5 hours first time i was trying to fix it 
