-
Notifications
You must be signed in to change notification settings - Fork 5
Tips & Tricks ‐ LoRA Settings
If you've found a great LoRA and you want to know settings it has been trained on, there's no need to ask the author! Sometimes it's possible to find its training settings in the model metadata. You just need to import it in the AUTOMATIC 1111
. Then open Lora
tab. Above some models you'll see info button. If it's there, then you can see the model internal metadata.
You can't just import it into Kohya SS
or anywhere else. But it's pretty easy to match these settings with Kohya SS
settings. As you could see, there's a lot of information: checkpoint, resolution, clip skip, number of training and regularisation images, number of steps, LoRA type, scheduler, learning rates and everything else. In general, there's everything you need! Just fill Kohya SS
settings one by one, and you'll be fine.
- Introduction
- Examples
- Dataset Preparation
- Model Training ‐ Introduction
- Model Training ‐ Basics
- Model Training ‐ Comparison - Introduction
Short Way
Long Way
- Model Training ‐ Comparison - [Growth Rate]
- Model Training ‐ Comparison - [Betas]
- Model Training ‐ Comparison - [Weight Decay]
- Model Training ‐ Comparison - [Bias Correction]
- Model Training ‐ Comparison - [Decouple]
- Model Training ‐ Comparison - [Epochs x Repeats]
- Model Training ‐ Comparison - [Resolution]
- Model Training ‐ Comparison - [Aspect Ratio]
- Model Training ‐ Comparison - [Batch Size]
- Model Training ‐ Comparison - [Network Rank]
- Model Training ‐ Comparison - [Network Alpha]
- Model Training ‐ Comparison - [Total Steps]
- Model Training ‐ Comparison - [Scheduler]
- Model Training ‐ Comparison - [Noise Offset]
- Model Training ‐ Comparison - [Min SNR Gamma]
- Model Training ‐ Comparison - [Clip Skip]
- Model Training ‐ Comparison - [Precision]
- Model Training ‐ Comparison - [Number of CPU Threads per Core]
- Model Training ‐ Comparison - [Checkpoint]
- Model Training ‐ Comparison - [Regularisation]
- Model Training ‐ Comparison - [Optimizer]