update ap.py

app.py

@@ -1,64 +1,86 @@
+import torch, os
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-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")
+from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, Settings
+from llama_index.core.memory import ChatMemoryBuffer
+from llama_index.llms.huggingface import HuggingFaceLLM
+from llama_index.embeddings.langchain import LangchainEmbedding
+from langchain.embeddings.huggingface import HuggingFaceEmbeddings
+from llama_index.core import PromptTemplate
 
+# ---------- 1. Load menu PDF ----------
+docs = SimpleDirectoryReader(input_files=["cafeoneeleven.pdf"]).load_data()
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# ---------- 2. LLM + embeddings ----------
+SYSTEM_PROMPT = """
+You are a friendly café assistant for Café Eleven. Your job is to:
+1. Greet the customer warmly
+2. Help them place their order
+3. Ask for pickup time
+4. Suggest add-ons/extras from our menu
+5. Confirm the complete order
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+Menu items are embedded in the document. Always:
+- Be polite and professional
+- Confirm order details clearly
+- Suggest popular combinations
+- Never make up items not in our menu
+"""
 
-    messages.append({"role": "user", "content": message})
+wrapper_prompt = PromptTemplate(
+    """<s>[INST] <<SYS>>
+{system_prompt}
+Current conversation:
+{chat_history}
+<</SYS>>
 
-    response = ""
+{query_str} [/INST]"""
+)
 
-    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
+llm = HuggingFaceLLM(
+    tokenizer_name="meta-llama/Llama-2-7b-chat-hf",
+    model_name="meta-llama/Llama-2-7b-chat-hf",
+    context_window=3900,
+    max_new_tokens=256,
+    generate_kwargs={"temperature": 0.2, "do_sample": True},
+    device_map="auto",
+    model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
+    system_prompt=SYSTEM_PROMPT,
+    query_wrapper_prompt=wrapper_prompt,
+)
 
-        response += token
-        yield response
+embed_model = LangchainEmbedding(
+    HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-base-v2")
+)
 
+Settings.llm = llm
+Settings.embed_model = embed_model
+Settings.chunk_size = 512
 
-"""
-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)",
-        ),
-    ],
+# ---------- 3. Build the chat engine ----------
+memory = ChatMemoryBuffer.from_defaults(token_limit=2000)
+index = VectorStoreIndex.from_documents(docs)
+chat_engine = index.as_chat_engine(
+    chat_mode="condense_plus_context",
+    memory=memory,
+    system_prompt=SYSTEM_PROMPT,
 )
 
+# ---------- 4. Gradio UI ----------
+with gr.Blocks(title="Café Eleven Chat") as demo:
+    gr.Markdown("## ☕ Café Eleven Ordering Assistant  \nType *quit* to end the chat.")
+    chatbot = gr.Chatbot()
+    user_txt = gr.Textbox(show_label=False, placeholder="Hi, I’d like a latte…")
+    clear = gr.Button("Clear")
+
+    def respond(message, chat_history):
+        if message.lower().strip() in {"quit", "exit", "done"}:
+            return "Thank you for your order! We'll see you soon.", chat_history
+        response = chat_engine.chat(message).response
+        chat_history.append((message, response))
+        return "", chat_history
+
+    user_txt.submit(respond, [user_txt, chatbot], [user_txt, chatbot])
+    clear.click(lambda: None, None, chatbot, queue=False)
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.queue(concurrency_count=3).launch()