Update app.py

app.py

@@ -1,58 +1,40 @@
-from transformers import pipeline
-
-# Load the pre-trained or fine-tuned model
-model_pipeline = pipeline("text-generation", model="your_model_name")
-
-# Function to generate responses
-def respond(message, history=[], max_tokens=100, temperature=0.8, top_p=0.9):
-    """
-    Generates a meaningful response while avoiding repetition.
-    
-    Parameters:
-    - message (str): User's input message.
-    - history (list): List of previous (user, model) interactions.
-    - max_tokens (int): Maximum number of tokens to generate.
-    - temperature (float): Controls randomness (higher = more creative).
-    - top_p (float): Controls nucleus sampling.
-
-    Returns:
-    - response (str): Generated response.
-    """
-
-    message = message.strip()
-    
-    # Prevent repetition by checking history
-    if history and message.lower() == history[-1][0].lower():
-        return "Tafadhali uliza swali tofauti."
-
-    # Generate response
-    response = model_pipeline(
-        message, 
-        max_length=max_tokens, 
-        temperature=temperature, 
-        top_p=top_p, 
-        pad_token_id=50256  # Ensures clean output for some models
-    )[0]["generated_text"].strip()
-
-    # Ensure response is meaningful and not just repeating input
-    if response.lower() == message.lower():
-        response = "Samahani, siwezi kujibu hilo kwa sasa."
-
-    # Store history for context (optional)
-    history.append((message, response))
-
-    return response
-
-# Example conversation loop
+import gradio as gr
+from huggingface_hub import InferenceClient
+
+client = InferenceClient("Futuresony/future_ai_12_10_2024.gguf")
+
+def format_alpaca_prompt(user_input, system_prompt):
+    """Formats input in Alpaca/LLaMA style"""
+    prompt = f"""{system_prompt}
+
+### Instruction:
+{user_input}
+
+### Response:
+"""
+    return prompt
+
+def respond(message, history, system_message, max_tokens, temperature, top_p):
+    formatted_prompt = format_alpaca_prompt(message, system_message)
+
+    response = client.text_generation(
+        formatted_prompt,  # ✅ Pass as a single string
+        max_new_tokens=max_tokens,  # ✅ Use max_new_tokens, not max_tokens
+        temperature=temperature,
+        top_p=top_p,
+    )
+
+    yield response  # ✅ Output the generated response
+
+demo = gr.ChatInterface(
+    respond,
+    additional_inputs=[
+        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
+        gr.Slider(minimum=1, maximum=250, value=128, step=1, label="Max new tokens"),  # ✅ Keep ≤250
+        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
+    ],
+)
+
 if __name__ == "__main__":
-    print("🤖 Chatbot Ready! Type 'quit' to exit.")
-    chat_history = []
-    
-    while True:
-        user_input = input("You: ")
-        if user_input.lower() == "quit":
-            print("Chatbot: Karibu tena!")
-            break
-
-        bot_response = respond(user_input, chat_history)
-        print(f"Chatbot: {bot_response}")
+    demo.launch()