Replies: 2 comments 1 reply
-
pick the 10 best images from the dataset, rename them to one unique unknown word, and train with them with the default settings, no captions, no regularization |
Beta Was this translation helpful? Give feedback.
0 replies
-
thanks for the reply! Is there any other way to keep the 100 instances instead of 10? is it possible ? |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi everyone
I need to train a graphic style on a data set of 100 images of a a particular artist. Can anyone tell me a goos setting to start? how many training steps for 100 instances images ? And how many regularisation images ( I don't intent to mix styles, just wanna create new images with style accuracy ) How important are captions in this case?
thanks in advance
Beta Was this translation helpful? Give feedback.
All reactions