Update app.py

app.py

@@ -1,10 +1,7 @@
 from huggingface_hub import InferenceClient
 import gradio as gr
 
-client = InferenceClient(
-    "mistralai/Mistral-7B-Instruct-v0.1"
-)
-
+client = InferenceClient("mistralai/Mistral-7B-Instruct-v0.1")
 
 def format_prompt(message, history):
   prompt = "<s>"
@@ -14,14 +11,11 @@ def format_prompt(message, history):
   prompt += f"[INST] {message} [/INST]"
   return prompt
 
-def generate(
-    prompt, history, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
-):
+def generate(prompt, history, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,):
     temperature = float(temperature)
     if temperature < 1e-2:
         temperature = 1e-2
     top_p = float(top_p)
-
     generate_kwargs = dict(
         temperature=temperature,
         max_new_tokens=max_new_tokens,
@@ -30,67 +24,23 @@ def generate(
         do_sample=True,
         seed=42,
     )
-
     formatted_prompt = format_prompt(prompt, history)
-
     stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
     output = ""
-
     for response in stream:
         output += response.token.text
         yield output
     return output
 
-
 additional_inputs=[
-    gr.Slider(
-        label="Temperature",
-        value=0.9,
-        minimum=0.0,
-        maximum=1.0,
-        step=0.05,
-        interactive=True,
-        info="Higher values produce more diverse outputs",
-    ),
-    gr.Slider(
-        label="Max new tokens",
-        value=256,
-        minimum=0,
-        maximum=1048,
-        step=64,
-        interactive=True,
-        info="The maximum numbers of new tokens",
-    ),
-    gr.Slider(
-        label="Top-p (nucleus sampling)",
-        value=0.90,
-        minimum=0.0,
-        maximum=1,
-        step=0.05,
-        interactive=True,
-        info="Higher values sample more low-probability tokens",
-    ),
-    gr.Slider(
-        label="Repetition penalty",
-        value=1.2,
-        minimum=1.0,
-        maximum=2.0,
-        step=0.05,
-        interactive=True,
-        info="Penalize repeated tokens",
-    )
+    gr.Slider(label="Temperature", value=0.9, minimum=0.0, maximum=1.0, step=0.05, interactive=True, info="Higher values produce more diverse outputs",),
+    gr.Slider(label="Max new tokens", value=256, minimum=0, maximum=1048, step=64, interactive=True, info="The maximum numbers of new tokens",),
+    gr.Slider(label="Top-p (nucleus sampling)",value=0.90,minimum=0.0,maximum=1,step=0.05,interactive=True,info="Higher values sample more low-probability tokens",),
+    gr.Slider(label="Repetition penalty",value=1.2,minimum=1.0,maximum=2.0,step=0.05,interactive=True,info="Penalize repeated tokens",)
 ]
 
-css = """
-  #mkd {
-    height: 200px; 
-    overflow: auto; 
-    border: 1px solid #ccc; 
-  }
-"""
-
+css = """#mkd {height: 200px; overflow: auto; border: 1px solid #ccc;}"""
 with gr.Blocks(css=css) as demo:
-  
     gr.ChatInterface(
         generate,
         additional_inputs=additional_inputs,       
@@ -151,35 +101,25 @@ with gr.Blocks(css=css) as demo:
     | 📈 Expected speedups with Flash Attention 2 | Upcoming update expected to bring speed improvements. | Keep an eye out for this update to benefit from performance gains. |
 
 # 🛠 Model Features and More 🛠
-
 ## Features
-
 - 🪟 Sliding Window Attention with 128K tokens span  
   - **Byline**: Increases model's understanding of context, resulting in more coherent and contextually relevant outputs.
-
 - 🚀 GQA for faster inference  
   - **Byline**: Speeds up the model inference time without sacrificing too much on accuracy.
-
 - 📝 Byte-fallback BPE tokenizer  
   - **Byline**: Allows the tokenizer to handle a wider variety of input text while keeping token size manageable.
-
 - 📜 License: Released under Apache 2.0 License  
   - **Byline**: Gives you a permissive free software license, allowing you freedom to use, modify, and distribute the code.
-
 ## Usage 📦
-
 - 📚 Available on Huggingface Hub  
   - **Byline**: Makes it easier to integrate the model into various projects.
-
 - 🐍 Python code snippets for easy setup  
   - **Byline**: Facilitates rapid development and deployment, especially useful for prototyping.
-
 - 📈 Expected speedups with Flash Attention 2  
   - **Byline**: Keep an eye out for this update to benefit from performance gains.
     """
     gr.Markdown(markdown)  
-    
-            
+               
     def SpeechSynthesis(result):
         documentHTML5='''
         <!DOCTYPE html>
@@ -207,9 +147,7 @@ with gr.Blocks(css=css) as demo:
         </html>
         '''
         gr.HTML(documentHTML5)
-        # components.html(documentHTML5, width=1280, height=1024)
-        #return result
+        
     SpeechSynthesis(markdown)
     
-    
 demo.queue().launch(debug=True)