Upscale latent space image
Browse files
README.md
CHANGED
|
@@ -166,7 +166,9 @@ EMA weight is not included and it's fp16.
|
|
| 166 |
If you want to continue training, use [`bp_1024_e10_ema.ckpt`](bp_1024_e10_ema.ckpt) which is the ema weight
|
| 167 |
and with fp32 precision.
|
| 168 |
|
| 169 |
-
For better performance, it is strongly recommended to use Clip skip (CLIP stop at last layers) 2.
|
|
|
|
|
|
|
| 170 |
|
| 171 |
## About the Model Name
|
| 172 |
|
|
|
|
| 166 |
If you want to continue training, use [`bp_1024_e10_ema.ckpt`](bp_1024_e10_ema.ckpt) which is the ema weight
|
| 167 |
and with fp32 precision.
|
| 168 |
|
| 169 |
+
For better performance, it is strongly recommended to use Clip skip (CLIP stop at last layers) 2. It's also recommended to use turn on
|
| 170 |
+
*Upscale latent space image when doing hires. fix* in the settings of [AUTOMATIC1111/stable-diffusion-webui](https://github.com/AUTOMATIC1111/stable-diffusion-webui)
|
| 171 |
+
which adds intricate details when using `Highres. fix`.
|
| 172 |
|
| 173 |
## About the Model Name
|
| 174 |
|