Update dataset/stable_cascade.py

dataset/stable_cascade.py

@@ -12,8 +12,9 @@ import torch.nn as nn
 import torch.utils.checkpoint
 import torchvision
 
-# Put this .py file into sd-scripts/library and run the training.
-# It will run 1 step and FP16 fix the model after.
+# Updated and fixed version of KBlueLeaf's FP16 fix script.
+# Put this .py file into sd-scripts/library (git clone https://github.com/kohya-ss/sd-scripts -b stable-cascade)
+# Run the training and it will run for 1 step, then it will FP16 fix the model and exit.
 
 fp16_fix_save_path = "/mnt/DataSSD/AI/SoteDiffusion/Wuerstchen3"