updated layout and added theme

app.py

@@ -12,13 +12,6 @@ tok = AutoTokenizer.from_pretrained("togethercomputer/RedPajama-INCITE-Chat-3B-v
 m = AutoModelForCausalLM.from_pretrained("togethercomputer/RedPajama-INCITE-Chat-3B-v1", torch_dtype=torch.float16)
 m = m.to('cuda:0')
 
-start_message = """<|SYSTEM|># RedPajamaAssistant
-- RedPajamaAssistant is A helpful and harmless Open Source AI Language Model developed by Stability and CarperAI.
-- RedPajamaAssistant is excited to be able to help the user, but will refuse to do anything that could be considered harmful to the user.
-- RedPajamaAssistant is more than just an information source, RedPajamaAssistant is also able to write poetry, short stories, and make jokes.
-- RedPajamaAssistant will refuse to participate in anything that could harm a human."""
-
-
 class StopOnTokens(StoppingCriteria):
     def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTensor, **kwargs) -> bool:
         #stop_ids = [[29, 13961, 31], [29, 12042, 31], 1, 0]
@@ -36,7 +29,7 @@ def user(message, history):
 
 
 
-def chat(curr_system_message, history):
+def chat(history, top_p, top_k, temperature): 
     # Initialize a StopOnTokens object
     stop = StopOnTokens()
 
@@ -53,32 +46,32 @@ def chat(curr_system_message, history):
         streamer=streamer,
         max_new_tokens=1024,
         do_sample=True,
-        top_p=0.95,
-        top_k=1000,
-        temperature=1.0,
+        top_p=top_p, #0.95,
+        top_k=top_k, #1000,
+        temperature=temperature, #1.0,
         num_beams=1,
         stopping_criteria=StoppingCriteriaList([stop])
     )
     t = Thread(target=m.generate, kwargs=generate_kwargs)
     t.start()
 
-    # print(history)
     # Initialize an empty string to store the generated text
     partial_text = ""
     for new_text in streamer:
-        print(new_text)
+        #print(new_text)
         if new_text != '<':
             partial_text += new_text
             history[-1][1] = partial_text.split('<bot>:')[-1]
-            # Yield an empty string to cleanup the message textbox and the updated conversation history
+            # Yield an empty string to clean up the message textbox and the updated conversation history
             yield history
     return partial_text
 
 
 title = """<h1 align="center">🔥RedPajama-INCITE-Chat-3B-v1</h1><br><h2 align="center">🏃‍♂️💨Streaming with Transformers & Gradio💪</h2>"""
-description = """<h3 align="center">This is a RedPajama Chat model fine-tuned using data from Dolly 2.0 and Open Assistant over the RedPajama-INCITE-Base-3B-v1 base model.</h3>"""
+description = """<br><br><h3 align="center">This is a RedPajama Chat model fine-tuned using data from Dolly 2.0 and Open Assistant over the RedPajama-INCITE-Base-3B-v1 base model.</h3>"""
+theme = gr.themes.Soft(primary_hue="red", secondary_hue= "red", neutral_hue="red",)
 
-with gr.Blocks() as demo:
+with gr.Blocks(theme=theme) as demo:
     gr.HTML(title)
     gr.HTML('''<center><a href="https://huggingface.co/spaces/ysharma/RedPajama-Chat-3B?duplicate=true"><img src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>Duplicate the Space to skip the queue and run in a private space</center>''')
     chatbot = gr.Chatbot().style(height=500)
@@ -91,17 +84,21 @@ with gr.Blocks() as demo:
                 submit = gr.Button("Submit")
                 stop = gr.Button("Stop")
                 clear = gr.Button("Clear")
-    system_msg = gr.Textbox(
-        start_message, label="System Message", interactive=False, visible=False)
+    
+    #Advanced options - top_p, temperature, top_k
+    with gr.Accordion("Advanced Options:", open=False):
+        top_p = gr.Slider( minimum=-0, maximum=1.0, value=0.95, step=0.05, interactive=True, label="Top-p",)
+        top_k = gr.Slider(minimum=0.0, maximum=1000, value=1000, step=1, interactive=True, label="Top-k", )
+        temperature = gr.Slider( minimum=-0, maximum=5.0, value=1.0, step=0.1, interactive=True, label="Temperature",)
 
     submit_event = msg.submit(fn=user, inputs=[msg, chatbot], outputs=[msg, chatbot], queue=False).then(
-        fn=chat, inputs=[system_msg, chatbot], outputs=[chatbot], queue=True)
+        fn=chat, inputs=[chatbot, top_p, top_k, temperature], outputs=[chatbot], queue=True)  
     submit_click_event = submit.click(fn=user, inputs=[msg, chatbot], outputs=[msg, chatbot], queue=False).then(
-        fn=chat, inputs=[system_msg, chatbot], outputs=[chatbot], queue=True)
+        fn=chat, inputs=[chatbot, top_p, top_k, temperature], outputs=[chatbot], queue=True)  
     stop.click(fn=None, inputs=None, outputs=None, cancels=[
                submit_event, submit_click_event], queue=False)
     clear.click(lambda: None, None, [chatbot], queue=False)
     gr.HTML(description)
 
 demo.queue(max_size=32, concurrency_count=2)
-demo.launch(debug=True)
+demo.launch(debug=True)