oO0
oO0
yes. Because searching files in external viewer is faster, then in web UI
**KonokoAz** Do you test them all, or do you know what is better for each case? Not all types that Kohya has, really working with all models
Seems like toml is temp file and contains data from JSON
Similar issue was with Stable Cascade. but at least it can redownload it without authenticate restriction
ppbrown, do EMA working good only with big epoch number?
File "...\OneTrainer\modules\ui\TrainUI.py", line 518, in __training_thread_function trainer.train() File "...\OneTrainer\modules\trainer\GenericTrainer.py", line 554, in train self.__execute_sample_during_training() File "....\OneTrainer\modules\trainer\GenericTrainer.py", line 190, in __execute_sample_during_training fun() File "....\OneTrainer\modules\trainer\GenericTrainer.py", line 537, in lambda: self.__sample_during_training(train_progress, train_device) File...
And we need merger and better captioner. Also motion lora training, VAE training, and converting 2B to 8B
does scripts or Simpletuner works on Windows?
404 for https://github.com/huggingface/diffusers@363d1ab7e24c5ed6c190abb00df66d9edb74383b