Update README.md
Browse files
README.md
CHANGED
|
@@ -8,6 +8,15 @@ tags:
|
|
| 8 |
---
|
| 9 |
# Files:
|
| 10 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
## **Loras/Mugen-Consistency-Test-Muon-000004.safetensors**
|
| 12 |
|
| 13 |
Consistency lora trained similarly to the ones below, but with Muon
|
|
|
|
| 8 |
---
|
| 9 |
# Files:
|
| 10 |
|
| 11 |
+
## **Loras/Mugen-lanzcos-test-000001.safetensors**
|
| 12 |
+
|
| 13 |
+
Very small lora trained after replacing the downscaling algorithm used in sd-scripts from `cv2.INTER_AREA` to `cv2.INTER_LANCZOS4`
|
| 14 |
+
|
| 15 |
+
Works to increase sharpness and fine details when setting the positive conditioning's original size greater than the target size's.
|
| 16 |
+
- ComfyUI nodes to make this easier [HERE](https://github.com/lRemixl/ComfyUI-sdxl-micro-conditioning)
|
| 17 |
+
|
| 18 |
+
My reasoning for this is in this [.docx file](Loras/documents/Reasoning%20behind%20lanzcos%20downscaling.docx)
|
| 19 |
+
|
| 20 |
## **Loras/Mugen-Consistency-Test-Muon-000004.safetensors**
|
| 21 |
|
| 22 |
Consistency lora trained similarly to the ones below, but with Muon
|