Update app.py

app.py

@@ -9,13 +9,17 @@ client = InferenceClient(model)
 # Embedded system prompt
 system_prompt_text = "You are a smart and helpful co-worker of Thailand based multi-national company PTT, and PTTEP. You help with any kind of request and provide a detailed answer to the question."
 
-def format_prompt_mixtral(message, history):
+# Read the content of the info.md file
+with open("info.md", "r") as file:
+    info_md_content = file.read()
+
+def format_prompt_mixtral(message, history, info_md_content):
     prompt = "<s>"
     if history:
         for user_prompt, bot_response in history:
             prompt += f"[INST] {user_prompt} [/INST]"
             prompt += f" {bot_response}</s> "
-    prompt += f"[INST] {message} [/INST]"
+    prompt += f"[INST] {info_md_content}\n\n{message} [/INST]"
     return prompt
 
 def chat_inf(prompt, history, seed, temp, tokens, top_p, rep_p):
@@ -31,7 +35,7 @@ def chat_inf(prompt, history, seed, temp, tokens, top_p, rep_p):
         seed=seed,
     )
 
-    formatted_prompt = format_prompt_mixtral(full_prompt, history)
+    formatted_prompt = format_prompt_mixtral(full_prompt, history, info_md_content)
     stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
     output = ""
     for response in stream:
@@ -51,8 +55,8 @@ def check_rand(inp, val):
     else:
         return gr.Slider(label="Seed", minimum=1, maximum=1111111111111111, value=int(val))
 
-with gr.Blocks() as app:
-    gr.HTML("""<center><h1 style='font-size:xx-large;'>Chatbot</h1><br><h3>running on Huggingface Inference </h3><br><h7>EXPERIMENTAL""")
+with gr.Blocks(auth=("Admin", "0112358")) as app:  # Add auth here
+    gr.HTML("""<center><h1 style='font-size:xx-large;'>Chatbot</h1><br><h3>running on Huggingface Inference Client</h3><br><h7>EXPERIMENTAL""")
     with gr.Row():
         chat = gr.Chatbot(height=500)
     with gr.Group():
@@ -72,8 +76,8 @@ with gr.Blocks() as app:
                     seed = gr.Slider(label="Seed", minimum=1, maximum=1111111111111111, step=1, value=rand_val)
                     tokens = gr.Slider(label="Max new tokens", value=3840, minimum=0, maximum=8000, step=64, interactive=True, visible=True, info="The maximum number of tokens")
                     temp = gr.Slider(label="Temperature", step=0.01, minimum=0.01, maximum=1.0, value=0.9)
-                    top_p = gr.Slider(label="Top-P", step=0.01, minimum=0.01, maximum=1.0, value=0.9)
-                    rep_p = gr.Slider(label="Repetition Penalty", step=0.1, minimum=0.1, maximum=2.0, value=1.0)
+                    top_p = gr.Slider(label="Top-P", step=0.01, minimum=0.01, maximum 1.0, value=0.9)
+                    rep_p = gr.Slider(label="Repetition Penalty", step=0.1, minimum=0.1, maximum 2.0, value=1.0)
 
     hid1 = gr.Number(value=1, visible=False)