better interface

app.py

@@ -12,6 +12,12 @@ model, _, transform = open_clip.create_model_and_transforms(
 )
 model.to(device)
 
+title="""<h1 align="center">CoCa: Contrastive Captioners</h1>"""
+description=(
+    """<br> An open source implementation of <strong>CoCa: Contrastive Captioners are Image-Text Foundation Models</strong> <a href=https://arxiv.org/abs/2205.01917>https://arxiv.org/abs/2205.01917.</a>
+    <br> Built using <a href=https://github.com/mlfoundations/open_clip>open_clip</a> with an effort from <a href=https://laion.ai/>LAION</a>. 
+    <br> For faster inference without waiting in queue, you may duplicate the space and upgrade to GPU in settings.<a href="https://huggingface.co/spaces/laion/CoCa?duplicate=true"> <img style="margin-top: 0em; margin-bottom: 0em" src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>"""
+)
 
 def output_generate(image):
     im = transform(image).unsqueeze(0).to(device)
@@ -34,17 +40,11 @@ def inference_caption(image, decoding_method="Beam search", rep_penalty=1.2, top
     return open_clip.decode(generated[0].detach()).split("<end_of_text>")[0].replace("<start_of_text>", "")
 
 paths = sorted(pathlib.Path("images").glob("*.jpg"))
-with gr.Blocks(
-    css="""
-    .message.svelte-w6rprc.svelte-w6rprc.svelte-w6rprc {font-size: 20px; margin-top: 20px}
-    #component-21 > div.wrap.svelte-w6rprc {height: 600px;}
-    """
-) as iface:
+with gr.Blocks() as iface:
     state = gr.State([])
 
-    # gr.Markdown(title)
-    # gr.Markdown(description)
-    # gr.Markdown(article)
+    gr.Markdown(title)
+    gr.Markdown(description)
 
     with gr.Row():
         with gr.Column(scale=1):
@@ -70,7 +70,7 @@ with gr.Blocks(
             top_p = gr.Slider(
                 minimum=0.0,
                 maximum=1.0,
-                value=1.0,
+                value=0.5,
                 step=0.1,
                 interactive=True,
                 label="Top p (used with nucleus sampling)",
@@ -84,7 +84,7 @@ with gr.Blocks(
                 value=20, label="Maximum Sequence Length", precision=0, interactive=True
             )
 
-        with gr.Column(scale=1.8):
+        with gr.Column(scale=1):
 
             with gr.Column():
                 caption_output = gr.Textbox(lines=1, label="Caption Output")
@@ -116,4 +116,9 @@ with gr.Blocks(
 #     article="""""",
 #     examples=[path.as_posix() for path in paths],
 # )
+
+    examples = gr.Examples(
+        examples=[path.as_posix() for path in paths],
+        inputs=[image_input],
+    )
 iface.launch()