This video is not helpful if you're looking to train a model to output a consistent character. Why would you include a variety of completely different characters (puppies) in a dataset? To what end? Any of these puppy images could've been prompted without a training model. 44 minutes worth of missed opportunity. 🙄
We'll see about getting a stream done for model training for consistent characters. Until then. you want 8-15 images of your face or the face you want to be put into a model at different angles. Try to make sure all your image are the same ratio as well.
@@leonardo.ai_community i tried everything. leonardo either has a problem reading my dataset properly, or something else went wrong. i really need a structured tutorial on this
Thank you so much. When will part 2 be posted.
Soon! I'm glad you enjoyed it!
This is very useful. Thank you!
Where are these sessions announced and how do you join them?
You can join the live events over on the Leo Discord server. There should be a link in the description
@@leonardo.ai_community Thanks!
We need SDXL training!
This video is not helpful if you're looking to train a model to output a consistent character. Why would you include a variety of completely different characters (puppies) in a dataset? To what end? Any of these puppy images could've been prompted without a training model. 44 minutes worth of missed opportunity. 🙄
We'll see about getting a stream done for model training for consistent characters. Until then. you want 8-15 images of your face or the face you want to be put into a model at different angles. Try to make sure all your image are the same ratio as well.
@@leonardo.ai_community i tried everything. leonardo either has a problem reading my dataset properly, or something else went wrong. i really need a structured tutorial on this