Update app.py

app.py

@@ -31,7 +31,8 @@ model.to("cuda:0")
 
 @spaces.GPU
 def bot_streaming(message, history):
-    print(message)
+    print(f'message is - {message}')
+    print(f'history is - {history}')
     if message["files"]:
         # message["files"][-1] is a Dict or just a string
         if type(message["files"][-1]) == dict:
@@ -47,14 +48,12 @@ def bot_streaming(message, history):
     try:
         if image is None:
             # Handle the case where image is None
-            gr.Error("You need to upload an image for LLaVA to work.")
+            gr.Error("You need to upload an image for FalconVLM to work.")
     except NameError:
         # Handle the case where 'image' is not defined at all
-        gr.Error("You need to upload an image for LLaVA to work.")
+        gr.Error("You need to upload an image for FalconVLM to work.")
 
-    #prompt = f"<|start_header_id|>user<|end_header_id|>\n\n<image>\n{message['text']}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
-    prompt = f"""User:<image>\n{message} Falcon:"""
-    # print(f"prompt: {prompt}")
+    prompt = f"""User:<image>\n{message['text']} Falcon:"""
     image = Image.open(image)
     inputs = processor(prompt, image, return_tensors='pt').to(0, torch.float16)
 
@@ -64,23 +63,13 @@ def bot_streaming(message, history):
     thread = Thread(target=model.generate, kwargs=generation_kwargs)
     thread.start()
 
-    text_prompt = f"<|start_header_id|>user<|end_header_id|>\n\n{message['text']}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"
-    # print(f"text_prompt: {text_prompt}")
-
     buffer = ""
-    time.sleep(0.5)
     for new_text in streamer:
         # find <|eot_id|> and remove it from the new_text
         if "<|eot_id|>" in new_text:
             new_text = new_text.split("<|eot_id|>")[0]
         buffer += new_text
-
-        # generated_text_without_prompt = buffer[len(text_prompt):]
-        generated_text_without_prompt = buffer
-        # print(generated_text_without_prompt)
-        time.sleep(0.06)
-        # print(f"new_text: {generated_text_without_prompt}")
-        yield generated_text_without_prompt
+        yield buffer
 
 
 chatbot=gr.Chatbot(placeholder=PLACEHOLDER,scale=1)
@@ -99,5 +88,6 @@ with gr.Blocks(fill_height=True, ) as demo:
     cache_examples=False,
     )
 
-demo.queue(api_open=False)
-demo.launch(show_api=False, share=False)
+demo.queue()
+demo.launch()
+