Add v-1-2 training summary
Browse files
README.md
CHANGED
@@ -18,7 +18,14 @@ extra_gated_fields:
|
|
18 |
---
|
19 |
|
20 |
# Stable Diffusion v1 Model Card
|
21 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
|
23 |
## Model Details
|
24 |
- **Developed by:** Robin Rombach, Patrick Esser
|
|
|
18 |
---
|
19 |
|
20 |
# Stable Diffusion v1 Model Card
|
21 |
+
|
22 |
+
Stable Diffusion is a latent text-to-image diffusion model capable of generating photo-realistic images given any text input.
|
23 |
+
|
24 |
+
The **Stable-Diffusion-v-1-2** checkpoint was initialized with the weights of the [Stable-Diffusion-v-1-1](https:/steps/huggingface.co/CompVis/stable-diffusion-v-1-1-original)
|
25 |
+
checkpoint and subsequently fine-tuned on 515,000 steps at resolution `512x512` on "laion-improved-aesthetics" (a subset of laion2B-en,
|
26 |
+
filtered to images with an original size `>= 512x512`, estimated aesthetics score `> 5.0`, and an estimated watermark probability `< 0.5`.
|
27 |
+
|
28 |
+
For more information, please refer to [Training](#training).
|
29 |
|
30 |
## Model Details
|
31 |
- **Developed by:** Robin Rombach, Patrick Esser
|