Update app.py

app.py

@@ -50,10 +50,16 @@ tokenizer, llava_model, image_processor, context_len = load_pretrained_model(
     args.load_4bit, 
     device=args.device)
 
+
+PLACEHOLDER = """
+<div style="padding: 30px; text-align: center; display: flex; flex-direction: column; align-items: center;">
+   <h1 style="font-size: 28px; margin-bottom: 2px; opacity: 0.55;">FinLLaVA</h1>
+   <p style="font-size: 18px; margin-bottom: 2px; opacity: 0.65;">FinLLaVA is a LLaVA model fine-tuned from FinLLaMA and CLIP by the Fin AI</p>
+</div>
+"""
+
 @spaces.GPU
 def bot_streaming(message, history):
-    print ("triggered")
-    print(message)
     image_file = None
     if message["files"]:
         if type(message["files"][-1]) == dict:
@@ -93,18 +99,17 @@ def bot_streaming(message, history):
     for new_text in streamer:
         buffer += new_text
         generated_text_without_prompt = buffer
-        time.sleep(0.06)
+        time.sleep(1.0)
         print (generated_text_without_prompt)
         yield generated_text_without_prompt
 
-chatbot = gr.Chatbot(scale=1)
+chatbot = gr.Chatbot(placeholder=PLACEHOLDER,scale=1)
 chat_input = gr.MultimodalTextbox(interactive=True, file_types=["image"], placeholder="Enter message or upload file...", show_label=False)
 with gr.Blocks(fill_height=True) as demo:
     gr.ChatInterface(
         fn=bot_streaming,
         title="FinLLaVA-8B Demo", 
         examples=[
-            {"text": "What is in this picture?", "files": ["http://images.cocodataset.org/val2017/000000039769.jpg"]},
             {"text": "What is the spending on Healthcare in July? A. 450 B. 600 C. 520 D. 510", "files": ["image_107.png"]},
             {"text": "If 2012 net periodic opeb cost increased at the same pace as the pension cost, what would the estimated 2013 cost be in millions? A. 14.83333 B. 12.5 C. 15.5 D. 13.5", "files": ["image_659.png"]},
         ],