Update generate_images_with_stable_diffusion.py (#1171)

* Update generate_images_with_stable_diffusion.py

* add: modified ipynb and md.

Author: sayakpaul

guides/ipynb/keras_cv/generate_images_with_stable_diffusion.ipynb

@@ -161,7 +161,6 @@
    ]
   },
   {
-   "attachments": {},
    "cell_type": "markdown",
    "metadata": {
     "colab_type": "text"
@@ -206,8 +205,8 @@
     "- A decoder, which turns the final 64x64 latent patch into a higher-resolution 512x512 image.\n",
     "\n",
     "First, your text prompt gets projected into a latent vector space by the text encoder,\n",
-    "which is simply a pretrained, frozen language model. Then that prompt vector is concatenate\n",
-    "to a randomly generated noise patch, which is repeatedly \"denoised\" by the decoder over a series\n",
+    "which is simply a pretrained, frozen language model. Then that prompt vector is concatenated\n",
+    "to a randomly generated noise patch, which is repeatedly \"denoised\" by the diffusion model over a series\n",
     "of \"steps\" (the more steps you run the clearer and nicer your image will be -- the default value is 50 steps).\n",
     "\n",
     "Finally, the 64x64 latent image is sent through the decoder to properly render it in high resolution.\n",
@@ -630,7 +629,7 @@
    "toc_visible": true
   },
   "kernelspec": {
-   "display_name": "Python 3.10.7 64-bit",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -644,7 +643,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.7"
+   "version": "3.8.2"
   },
   "vscode": {
    "interpreter": {
@@ -653,5 +652,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 0
+ "nbformat_minor": 1
 }

guides/keras_cv/generate_images_with_stable_diffusion.py

@@ -126,8 +126,8 @@ def plot_images(images):
 - A decoder, which turns the final 64x64 latent patch into a higher-resolution 512x512 image.
 
 First, your text prompt gets projected into a latent vector space by the text encoder,
-which is simply a pretrained, frozen language model. Then that prompt vector is concatenate
-to a randomly generated noise patch, which is repeatedly "denoised" by the decoder over a series
+which is simply a pretrained, frozen language model. Then that prompt vector is concatenated
+to a randomly generated noise patch, which is repeatedly "denoised" by the diffusion model over a series
 of "steps" (the more steps you run the clearer and nicer your image will be -- the default value is 50 steps).
 
 Finally, the 64x64 latent image is sent through the decoder to properly render it in high resolution.

guides/md/keras_cv/generate_images_with_stable_diffusion.md

@@ -154,8 +154,8 @@ This gives rise to the Stable Diffusion architecture. Stable Diffusion consists
 - A decoder, which turns the final 64x64 latent patch into a higher-resolution 512x512 image.
 
 First, your text prompt gets projected into a latent vector space by the text encoder,
-which is simply a pretrained, frozen language model. Then that prompt vector is concatenate
-to a randomly generated noise patch, which is repeatedly "denoised" by the decoder over a series
+which is simply a pretrained, frozen language model. Then that prompt vector is concatenated
+to a randomly generated noise patch, which is repeatedly "denoised" by the diffusion model over a series
 of "steps" (the more steps you run the clearer and nicer your image will be -- the default value is 50 steps).
 
 Finally, the 64x64 latent image is sent through the decoder to properly render it in high resolution.