ProGamerGov
commited on
Commit
•
4352068
1
Parent(s):
de3585c
Add inspiration to ReadMe
Browse files
README.md
CHANGED
@@ -9,7 +9,7 @@ datasets:
|
|
9 |
|
10 |
**Object-Taped-To-Wall-Diffusion**
|
11 |
|
12 |
-
This fine-tuned Stable Diffusion v1.5 model was trained for 2000 iterations with a batch size of 4, on a selection of photos of things taped to wall. Training was performed using [ShivamShrirao/diffusers](https://github.com/ShivamShrirao/diffusers) with full precision, prior-preservation loss, the train-text-encoder feature, and the new [1.5 MSE VAE from Stability AI](https://huggingface.co/stabilityai/sd-vae-ft-mse). A total of 2100 regularization / class images were used from [here](https://huggingface.co/datasets/ProGamerGov/StableDiffusion-v1-5-Regularization-Images). Regularization images were generated using the prompt "artwork style" with 50 DPM++ 2S a Karras steps and a CFG of 7, using the MSE VAE. A negative prompt of "text" was also used for this dataset.
|
13 |
|
14 |
Use the tokens **ttw style** in your prompts for the effect. Note that the effect also appears to occur at a much weaker strength on prompts that steer the output towards specific artistic styles.
|
15 |
|
@@ -28,3 +28,6 @@ Example images were generated with the v1 2000 iteration model using DPM++ 2S a
|
|
28 |
```
|
29 |
ttw style, <object> taped to wall
|
30 |
```
|
|
|
|
|
|
9 |
|
10 |
**Object-Taped-To-Wall-Diffusion**
|
11 |
|
12 |
+
This fine-tuned Stable Diffusion v1.5 model was trained for 2000 iterations with a batch size of 4, on a selection of photos of things taped to a wall. Training was performed using [ShivamShrirao/diffusers](https://github.com/ShivamShrirao/diffusers) with full precision, prior-preservation loss, the train-text-encoder feature, and the new [1.5 MSE VAE from Stability AI](https://huggingface.co/stabilityai/sd-vae-ft-mse). A total of 2100 regularization / class images were used from [here](https://huggingface.co/datasets/ProGamerGov/StableDiffusion-v1-5-Regularization-Images). Regularization images were generated using the prompt "artwork style" with 50 DPM++ 2S a Karras steps and a CFG of 7, using the MSE VAE. A negative prompt of "text" was also used for this dataset.
|
13 |
|
14 |
Use the tokens **ttw style** in your prompts for the effect. Note that the effect also appears to occur at a much weaker strength on prompts that steer the output towards specific artistic styles.
|
15 |
|
28 |
```
|
29 |
ttw style, <object> taped to wall
|
30 |
```
|
31 |
+
|
32 |
+
|
33 |
+
This model was inspired by the 2019 art piece [*Comedian* by Italian artist Maurizio Cattelan](https://en.wikipedia.org/wiki/Comedian_(artwork\)), where a banana was duct taped to a wall.
|