Update README.md
Browse files
README.md
CHANGED
|
@@ -95,6 +95,16 @@ When you pay attention to above image you will see now it is way more consistent
|
|
| 95 |
|
| 96 |
# Best Checkpoint And Conclusion
|
| 97 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 98 |
# Tutorials To Train Your Style
|
| 99 |
|
| 100 |
1 : https://youtu.be/bupRePUOA18
|
|
|
|
| 95 |
|
| 96 |
# Best Checkpoint And Conclusion
|
| 97 |
|
| 98 |
+
When inconsistent dataset was used, training with captions yielded way better results.
|
| 99 |
+
|
| 100 |
+
However, when training made with a consistent dataset, no captions yielded better and more consistent results with early epochs.
|
| 101 |
+
|
| 102 |
+
Thus I concluded that, epoch 75 of no-captions dataset is best checkpoint
|
| 103 |
+
|
| 104 |
+
Here below comparison images for fixed dataset
|
| 105 |
+
|
| 106 |
+
[Fixed-Consistent-Training-Dataset-No-Captions-Only-Grid.jpg](https://huggingface.co/MonsterMMORPG/3D-Cartoon-Style-FLUX/resolve/main/Fixed-Consistent-Training-Dataset-No-Captions-Only-Grid.jpg)
|
| 107 |
+
|
| 108 |
# Tutorials To Train Your Style
|
| 109 |
|
| 110 |
1 : https://youtu.be/bupRePUOA18
|