Update app.py

app.py

@@ -16,19 +16,20 @@ is_colab = utils.is_google_colab()
 colab_instruction = "" if is_colab else """
 <p>You can skip the queue using Colab: <a href="https://colab.research.google.com/gist/ChenWu98/0aa4fe7be80f6b45d3d055df9f14353a/copy-of-fine-tuned-diffusion-gradio.ipynb"><img data-canonical-src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab" src="https://colab.research.google.com/assets/colab-badge.svg"></a></p>"""
 
-model_id_or_path = "CompVis/stable-diffusion-v1-4"
-if is_colab:
-    scheduler = DDIMScheduler.from_config(model_id_or_path, subfolder="scheduler")
-    pipe = CycleDiffusionPipeline.from_pretrained(model_id_or_path, scheduler=scheduler)
-else:
-    import streamlit as st
-    scheduler = DDIMScheduler.from_config(model_id_or_path, use_auth_token=st.secrets["USER_TOKEN"], subfolder="scheduler")
-    torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
-    pipe = CycleDiffusionPipeline.from_pretrained(model_id_or_path, use_auth_token=st.secrets["USER_TOKEN"], scheduler=scheduler, torch_dtype=torch_dtype)
-tokenizer = pipe.tokenizer
-
-if torch.cuda.is_available():
-    pipe = pipe.to("cuda")
+if True:
+    model_id_or_path = "CompVis/stable-diffusion-v1-4"
+    if is_colab:
+        scheduler = DDIMScheduler.from_config(model_id_or_path, subfolder="scheduler")
+        pipe = CycleDiffusionPipeline.from_pretrained(model_id_or_path, scheduler=scheduler)
+    else:
+        import streamlit as st
+        scheduler = DDIMScheduler.from_config(model_id_or_path, use_auth_token=st.secrets["USER_TOKEN"], subfolder="scheduler")
+        torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+        pipe = CycleDiffusionPipeline.from_pretrained(model_id_or_path, use_auth_token=st.secrets["USER_TOKEN"], scheduler=scheduler, torch_dtype=torch_dtype)
+    tokenizer = pipe.tokenizer
+
+    if torch.cuda.is_available():
+        pipe = pipe.to("cuda")
 
 device_print = "GPU 🔥" if torch.cuda.is_available() else "CPU 🥶"
 device = "cuda" if torch.cuda.is_available() else "cpu"
@@ -294,8 +295,19 @@ with gr.Blocks(css=css) as demo:
               <p>
               <b>Quick start</b>: <br>
               1. Click one row of Examples at the end of this page. It will fill all inputs needed. <br>
-              2. Click the "Edit" button. <br>
+              2. Click the "Run CycleDiffusion" button. <br>
               </p>
+              <p>
+              {colab_instruction}
+               Running on <b>{device_print}</b>{(" in a <b>Google Colab</b>." if is_colab else "")}
+              </p>
+            </div>
+        """
+    )
+    with gr.Accordion("See Details", open=False):
+        gr.HTML(
+            f"""
+            <div class="cycle-diffusion-div">
               <p>
                 <b>How to use:</b> <br>
                 1. Upload an image. <br>
@@ -304,7 +316,7 @@ with gr.Blocks(css=css) as demo:
                 4. Select the strength (smaller strength means better content preservation). <br>
                 5 (optional). Configurate Cross Attention Control options (e.g., CAC type, cross replace steps, self replace steps). <br>
                 6 (optional). Configurate other options (e.g., image size, inference steps, random seed). <br>
-                7. Click the "Edit" button. <br>
+                7. Click the "Run CycleDiffusion" button. <br>
               </p>
               <p>
                 <b>Notes:</b> <br>
@@ -318,13 +330,9 @@ with gr.Blocks(css=css) as demo:
               1. 30s on A10G. <br>
               2. 90s on T4. <br>
               </p>
-              <p>
-              {colab_instruction}
-               Running on <b>{device_print}</b>{(" in a <b>Google Colab</b>." if is_colab else "")}
-              </p>
             </div>
         """
-    )
+        )
     with gr.Row():
 
         with gr.Column(scale=55):