Update app.py

app.py

@@ -1,63 +1,67 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from model_setup import load_model
+from response_generator import generate_response
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+model, tokenizer = load_model()
 
+def generate_response_stream(user_input, chat_history):
+    try:
+        prompt = user_input
+        text = generate_response(prompt, model, tokenizer)
+        chat_history.append((prompt, ""))
+        words = text.split()
+        for i, word in enumerate(words):
+            chat_history[-1] = (prompt, " ".join(words[:i + 1]))
+            yield chat_history
+    except Exception as e:
+        chat_history.append(("Error", f"Error: {str(e)}"))
+        yield chat_history
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+custom_css = """
+#title {
+    font-size: 3em;
+    text-align: center;
+    font-weight: bold;
+    margin-bottom: 20px;
+    margin-top: 20px;
+    color: #333;
+}
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+#interface {
+    background-color: #f5f5f5;
+    padding: 30px;
+    border-radius: 15px;
+    width: 80%;
+    max-width: 1200px;
+    margin: auto;
+    box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
+}
 
-    messages.append({"role": "user", "content": message})
+#chatbot {
+    min-height: auto;
+    max-height: none;
+    overflow-y: visible;
+    border: 0.5px solid #ddd;
+    padding: 15px;
+    background-color: #ffffff;
+    border-radius: 10px;
+    line-height: 1.5;
+    font-size: 1.6em;
+}
 
-    response = ""
+#chatbot p {
+    margin: 0;
+}
+"""
 
-    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
+with gr.Blocks(css=custom_css) as iface:
+    gr.Markdown("<h1 id='title'>Hệ Thống Hỏi Đáp Y Tế VSS AI</h1>")
+    gr.Markdown("Nhập câu hỏi của bạn vào ô bên dưới và nhận phản hồi lại từ hệ thống của chúng tôi.")
+    chatbot = gr.Chatbot(elem_id="chatbot", label="Trò chuyện")
+    user_input = gr.Textbox(label="Nhập câu hỏi của bạn tại đây", placeholder="Ví dụ: Các vấn đề bạn cần hỗ trợ là gì?")
+    submit_button = gr.Button("Gửi câu hỏi")
 
-        response += token
-        yield response
+    user_input.submit(generate_response_stream, inputs=[user_input, gr.State([])], outputs=[chatbot])
+    submit_button.click(generate_response_stream, inputs=[user_input, gr.State([])], outputs=[chatbot])
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        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)",
-        ),
-    ],
-)
-
-
-if __name__ == "__main__":
-    demo.launch()
+iface.launch(share=True)