Update app.py

app.py

@@ -1,65 +1,87 @@
 
-from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, ServiceContext
-from llama_index.llms.huggingface import HuggingFaceLLM
-import torch
-import gradio as gr
+import os
+from getpass import getpass
+
+os.environ["OPENAI_API_KEY"] = getpass("Enter your OpenAI API key: ")
+
+
+
+
+from llama_index.llms.openai import OpenAI
+from llama_index.embeddings.openai import OpenAIEmbedding
 from llama_index.core import Settings
-from langchain_community.embeddings import HuggingFaceEmbeddings
-from llama_index.core import ServiceContext
-# from langchain.embeddings.huggingface import HuggingFaceEmbeddings
-
-documents = SimpleDirectoryReader('files').load_data()
-
-system_prompt="""
-You are a Q&A assistant. Your goal is to answer questions as
-accurately as possible based on the instructions and context provided.
-"""
-
-embed_model= HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-base-v2")
-
-llm = HuggingFaceLLM(
-    context_window=4096,
-    max_new_tokens=256,
-    generate_kwargs={"temperature": 0.1, "do_sample": True},
-    system_prompt=system_prompt,
-    tokenizer_name="anasmkh/new_customized_llama3.1_8b",
-    model_name="anasmkh/new_customized_llama3.1_8b",
-    device_map="auto",
-    model_kwargs={"torch_dtype": torch.float16 }
-)
 
+Settings.llm = OpenAI(model="gpt-3.5-turbo",temperature=0.4)
+Settings.embed_model = OpenAIEmbedding(model="text-embedding-ada-002")
+
+from llama_index.core import SimpleDirectoryReader
+
+documents = SimpleDirectoryReader("/content/new_files").load_data()
 
+from llama_index.core import VectorStoreIndex, StorageContext
+from llama_index.vector_stores.qdrant import QdrantVectorStore
+import qdrant_client
 
+client = qdrant_client.QdrantClient(
+    location=":memory:",
+)
+
+vector_store = QdrantVectorStore(
+    collection_name = "paper",
+    client=client,
+    enable_hybrid=True,
+    batch_size=20,
+)
 
-Settings.llm = llm
-Settings.embed_model =embed_model
-# Settings.node_parser = SentenceSplitter(chunk_size=512, chunk_overlap=20)
-Settings.num_output = 250
-Settings.context_window = 3900
+storage_context = StorageContext.from_defaults(vector_store=vector_store)
 
 index = VectorStoreIndex.from_documents(
-    documents, embed_model=embed_model
+    documents,
+    storage_context=storage_context,
 )
 
-query_engine = index.as_query_engine(llm=llm)
+query_engine = index.as_query_engine(
+    vector_store_query_mode="hybrid"
+)
+
+from llama_index.core.memory import ChatMemoryBuffer
+
+memory = ChatMemoryBuffer.from_defaults(token_limit=3000)
+
+chat_engine = index.as_chat_engine(
+    chat_mode="context",
+    memory=memory,
+    system_prompt=(
+        "You are an AI assistant who answers the user questions"
+    ),
+)
+
+import gradio as gr
+def chat_with_ai(user_input, chat_history):
+    response = chat_engine.chat(user_input)
+
+    chat_history = chat_history + [(user_input, str(response))]
+
+    return chat_history, ""
+
+
+def gradio_chatbot():
+    with gr.Blocks() as demo:
+        gr.Markdown("# Gradio Chat Interface for LlamaIndex")
 
+        chatbot = gr.Chatbot(label="LlamaIndex Chatbot")
+        user_input = gr.Textbox(
+            placeholder="Ask a question...", label="Enter your question"
+        )
 
+        submit_button = gr.Button("Send")
 
+        chat_history = gr.State([])
 
-def chat(message, history):
-    history = history or []
-    history.append({"role": "user", "content": message})
-    response=query_engine.query(message)
-    # response = generator(history)[-1]["generated_text"]
-    history.append({"role": "assistant", "content": response})
-    return history
+        submit_button.click(chat_with_ai, inputs=[user_input, chat_history], outputs=[chatbot, user_input])
 
-with gr.Blocks() as demo:
-    chatbot = gr.Chatbot()
-    message = gr.Textbox()
-    clear = gr.ClearButton([message, chatbot])
+        user_input.submit(chat_with_ai, inputs=[user_input, chat_history], outputs=[chatbot, user_input])
 
-    message.submit(chat, [message, chatbot], chatbot)
-    clear.click(lambda: None, None, chatbot, queue=False)
+    return demo
 
-demo.launch()
+gradio_chatbot().launch(debug=True)