Update app.py

app.py

@@ -1,26 +1,77 @@
 import gradio as gr
-from transformers import AutoModelForCausalLM, AutoTokenizer
-import torch
-
-# Load the model and tokenizer
-model_name = "indonlp/cendol-llama2-7b-chat"  
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
-
-# Define the chatbot function
-def chatbot(input_text):
-    inputs = tokenizer(input_text, return_tensors="pt")
-    with torch.no_grad():
-        outputs = model.generate(inputs.input_ids, max_length=1000, do_sample=True)
-    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return response
-
-# Create the Gradio interface
-interface = gr.ChatInterface(
-    fn=chatbot,
-    title="Cendol Chatnot",
-    description="Ask any coding-related questions and get helpful responses."
+from transformers import pipeline
+from huggingface_hub import InferenceClient
+import requests
+from bs4 import BeautifulSoup
+
+# Initialize the text generation pipeline
+pipe = pipeline("text-generation", model="microsoft/Phi-3-mini-4k-instruct", trust_remote_code=True)
+client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+
+def web_search(query):
+    # Simulate a web search using Google
+    response = requests.get(f"https://www.google.com/search?q={query}")
+    soup = BeautifulSoup(response.text, "html.parser")
+    results = []
+    for g in soup.find_all('div', class_='BNeawe vvjwJb AP7Wnd'):
+        results.append(g.get_text())
+    return results
+
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+):
+    messages = [{"role": "system", "content": system_message}]
+
+    for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
+
+    # Check if message is a search request
+    if "search:" in message.lower():
+        search_query = message.split("search:", 1)[1].strip()
+        search_results = web_search(search_query)
+        response = "\n".join(search_results[:5])  # Return top 5 search results
+    else:
+        messages.append({"role": "user", "content": message})
+        response = ""
+        for message in client.chat_completion(
+            messages,
+            max_tokens=max_tokens,
+            stream=True,
+            temperature=temperature,
+            top_p=top_p,
+        ):
+            token = message.choices[0].delta.content
+            response += token
+            yield response
+
+    yield response
+
+
+demo = gr.ChatInterface(
+    respond,
+    title="INDONESIAN CHATBOT"
+    additional_inputs=[
+        gr.Textbox(value="You are a friendly AI Assistens Speak in indonesian", label="System message", visible=False),
+        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+        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)",
+        ),
+    ],
 )
 
-# Launch the web UI
-interface.launch(share=True)
+
+if __name__ == "__main__":
+    demo.launch()