Update README.md
Browse files
README.md
CHANGED
|
@@ -9,6 +9,7 @@ Put the yaml file with model because it is sd2.1 v_prediction model.
|
|
| 9 |
Model is the difference from the original checkpoint created by [extract_controlnet_diff.py](https://github.com/Mikubill/sd-webui-controlnet/blob/main/extract_controlnet_diff.py).
|
| 10 |
|
| 11 |
# diff_control_wd15beta2_canny.safetensors
|
|
|
|
| 12 |
+ target model: https://huggingface.co/waifu-diffusion/wd-1-5-beta2
|
| 13 |
+ dataset: 11k 1girl images
|
| 14 |
+ epoch: 2
|
|
@@ -17,6 +18,7 @@ Model is the difference from the original checkpoint created by [extract_control
|
|
| 17 |
+ amp:bfloat16
|
| 18 |
|
| 19 |
# diff_control_wd15beta2_pose.safetensors
|
|
|
|
| 20 |
+ target model: https://huggingface.co/waifu-diffusion/wd-1-5-beta2
|
| 21 |
+ dataset: 14k umamusume images. **Therefore, the model may have a bias**
|
| 22 |
+ epoch: 5 (The “sudden convergence phenomenon” occurred at the fourth epoch)
|
|
|
|
| 9 |
Model is the difference from the original checkpoint created by [extract_controlnet_diff.py](https://github.com/Mikubill/sd-webui-controlnet/blob/main/extract_controlnet_diff.py).
|
| 10 |
|
| 11 |
# diff_control_wd15beta2_canny.safetensors
|
| 12 |
+
It seems pretty good.
|
| 13 |
+ target model: https://huggingface.co/waifu-diffusion/wd-1-5-beta2
|
| 14 |
+ dataset: 11k 1girl images
|
| 15 |
+ epoch: 2
|
|
|
|
| 18 |
+ amp:bfloat16
|
| 19 |
|
| 20 |
# diff_control_wd15beta2_pose.safetensors
|
| 21 |
+
It seems a little subtle.
|
| 22 |
+ target model: https://huggingface.co/waifu-diffusion/wd-1-5-beta2
|
| 23 |
+ dataset: 14k umamusume images. **Therefore, the model may have a bias**
|
| 24 |
+ epoch: 5 (The “sudden convergence phenomenon” occurred at the fourth epoch)
|