Update README.md

README.md

@@ -11,25 +11,23 @@ inference: false
 
 # waifu-diffusion - Diffusion for Weebs
 
-waifu-diffusion is a latent text-to-image diffusion model that has been conditioned on high-quality anime images through [Textual Inversion](https://github.com/rinongal/textual_inversion).
+waifu-diffusion is a latent text-to-image diffusion model that has been conditioned on high-quality anime images through fine-tuning on high quality anime images.
 
 ## Model Description
 
-The model originally used for fine-tuning is [Stable Diffusion V1-4](https://huggingface.co/CompVis/stable-diffusion-v1-4), which is a latent image diffusion model trained on [LAION2B-en](https://huggingface.co/datasets/laion/laion2B-en).
+The model used for fine-tuning is [Stable Diffusion V1-4](https://huggingface.co/CompVis/stable-diffusion-v1-4), which is a latent text-to-image diffusion model trained on [LAION2B-en](https://huggingface.co/datasets/laion/laion2B-en).
 
-The current model is based from [Yasu Seno](https://twitter.com/naclbbr)'s [TrinArt Stable Diffusion](https://huggingface.co/naclbit/trinart_stable_diffusion) which has been fine-tuned on 30,000 high-resolution manga/anime-style images for 3.5 epochs.
+The current model is fine-tuned from 56 thousand images from Danbooru selected with an aesthetic score greater than `6.0`.
 
 With [Textual Inversion](https://github.com/rinongal/textual_inversion), the embeddings for the text encoder has been trained to align more with anime-styled images, reducing excessive prompting.
 
 ## Training Data & Annotative Prompting
 
-The data used for Textual Inversion has come from a random sample of 25k Danbooru images, which were then filtered based on [CLIP Aesthetic Scoring](https://github.com/christophschuhmann/improved-aesthetic-predictor) where only images with an aesthetic score greater than `6.0` were used.
-
-Then, the embeddings were further tuned on a smaller subset of 2k higher quality aesthetic images which had an aesthetic score greater than `6.0` and featured diverse subjects, backgrounds, and compositions.
+The data used for fine-tuning has come from a random sample of 56k Danbooru images, which were filtered based on [CLIP Aesthetic Scoring](https://github.com/christophschuhmann/improved-aesthetic-predictor) where only images with an aesthetic score greater than `6.0` were used.
 
 ## Downstream Uses
 
-This model can be used for entertainment purposes and as a generative art assistant.
+This model can be used for entertainment purposes and as a generative art assistant. The EMA model can be used for additional fine-tuning.
 
 ## Example Code
 
@@ -54,9 +52,7 @@ image.save("reimu_hakurei.png")
 
 ## Team Members and Acknowledgements
 
-This project would not have been possible without the incredible work by the [CompVis Researchers](https://ommer-lab.com/) and the author of the original finetuned model that this work was based upon, [Yasu Seno](https://twitter.com/naclbbr)!
-
-Additionally, the methods presented in the [Textual Inversion](https://github.com/rinongal/textual_inversion) repo was an incredible help.
+This project would not have been possible without the incredible work by the [CompVis Researchers](https://ommer-lab.com/).
 
 - [Anthony Mercurio](https://github.com/harubaru)
 - [Salt](https://github.com/sALTaccount/)