Update README.md

README.md

@@ -9,7 +9,7 @@ tags:
 - stable-diffusion-diffusers
 base_model: stabilityai/stable-diffusion-2-1-base
 inference: true
-instance_prompt: disney
+instance_prompt: disney style
 ---
 
 <!-- This model card has been generated automatically according to the information the training script had access to. You
@@ -18,12 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # DreamBooth - lavaman131/cartoonify
 
-This is a dreambooth model derived from stabilityai/stable-diffusion-2-1-base. The weights were trained on disney using [DreamBooth](https://dreambooth.github.io/).
-You can find some example images in the following. 
+This is a dreambooth model derived from runwayml/stable-diffusion-v1-5 with fine-tuning of the text encoder. The weights were trained from a popular animation studio using [DreamBooth](https://dreambooth.github.io/). Use the tokens **_disney style_** in your prompts for the effect.
 
+You can find some example images below:
 
-
-DreamBooth for the text encoder was enabled: True.
+![](./images/king.png)
 
 
 ## Intended uses & limitations
@@ -31,13 +30,19 @@ DreamBooth for the text encoder was enabled: True.
 #### How to use
 
 ```python
-# TODO: add an example code snippet for running this diffusion pipeline
+# basic usage
+repo_id = "lavaman131/cartoonify"
+torch_dtype = torch.float16 if device.type in ["mps", "cuda"] else torch.float32
+pipeline = StableDiffusionPipeline.from_pretrained(repo_id, torch_dtype=torch_dtype).to(device)
+image = pipeline("PROMPT GOES HERE").images[0]
 ```
 
 #### Limitations and bias
 
-[TODO: provide examples of latent issues and potential remediations]
+As with any diffusion model, playing around with the prompt and classifier-free guidance parameter is required until you get the results you want. For additional safety in image generation, we use the Stable Diffusion safety checker.
 
 ## Training details
 
-[TODO: describe the data used to train the model]
+The model was fine-tuned for 3500 steps on around 200 images of modern Disney characters, backgrounds, and animals. The ratios for each were 70%, 20%, and 10% respectively on an RTX A5000 GPU (24GB VRAM).
+
+The training code used can be found [here](https://github.com/huggingface/diffusers/blob/main/examples/dreambooth/train_dreambooth.py). The regularization images used for training can be found [here](https://github.com/aitrepreneur/SD-Regularization-Images-Style-Dreambooth/tree/main/style_ddim).