Update README.md
Browse files
README.md
CHANGED
|
@@ -1,5 +1,9 @@
|
|
| 1 |
proof of concept of a failed project
|
| 2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
how you can load this into a diffusers-based notebook like [Doohickey](https://github.com/aicrumb/doohickey) might look something like this
|
| 4 |
|
| 5 |
```
|
|
@@ -12,3 +16,8 @@ if len(stable_inversion)>1:
|
|
| 12 |
```
|
| 13 |
|
| 14 |
it was trained on 1024 images matching the 'genshin_impact' tag on safebooru, epochs 1 and 2 had the model being fed the full captions, epoch 3 had 50% of the tags in the caption, and epoch 4 had 25% of the tags in the caption. Learning rate was 1e-3 and the loss curve looked like this 
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
proof of concept of a failed project
|
| 2 |
|
| 3 |
+
a notebook for producing your own "stable inversions" is included in this repo but I wouldn't recommend doing so (they suck). It works on Colab free tier though.
|
| 4 |
+
|
| 5 |
+
[link to notebook for you to download](https://huggingface.co/crumb/genshin-stable-inversion/blob/main/stable_inversion%20(1).ipynb)
|
| 6 |
+
|
| 7 |
how you can load this into a diffusers-based notebook like [Doohickey](https://github.com/aicrumb/doohickey) might look something like this
|
| 8 |
|
| 9 |
```
|
|
|
|
| 16 |
```
|
| 17 |
|
| 18 |
it was trained on 1024 images matching the 'genshin_impact' tag on safebooru, epochs 1 and 2 had the model being fed the full captions, epoch 3 had 50% of the tags in the caption, and epoch 4 had 25% of the tags in the caption. Learning rate was 1e-3 and the loss curve looked like this 
|
| 19 |
+
|
| 20 |
+
Samples from this finetuned inversion for the prompt "beidou_(genshin_impact)"
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
|