llama index upgrade version migration and ollama chat backend

app.py

@@ -4,27 +4,31 @@ import openai
 import os
 from dotenv import load_dotenv
 import phoenix as px
+
 import llama_index
-from llama_index import Prompt, ServiceContext, VectorStoreIndex, SimpleDirectoryReader
-from llama_index.chat_engine.types import ChatMode
-from llama_index.llms import ChatMessage, MessageRole
+from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, Settings
+from llama_index.core.chat_engine.types import ChatMode
+from llama_index.core.llms import ChatMessage, MessageRole
+from llama_index.core.node_parser import SentenceSplitter
+from llama_index.core.extractors import TitleExtractor
+from llama_index.core.ingestion import IngestionPipeline
 from llama_index.vector_stores.qdrant import QdrantVectorStore
-from llama_index.text_splitter import SentenceSplitter
-from llama_index.extractors import TitleExtractor
-from llama_index.ingestion import IngestionPipeline
+
+from llama_index.llms.ollama import Ollama
+
 from chat_template import CHAT_TEXT_QA_PROMPT, TEXT_QA_SYSTEM_PROMPT
 from schemas import ChatbotVersion, ServiceProvider
 from chatbot import Chatbot, IndexBuilder
 from custom_io import MarkdownReader, UnstructuredReader, default_file_metadata_func
 from qdrant import client as qdrantClient
-from llama_index import set_global_service_context
+from llama_index.core import set_global_service_context
 
 from service_provider_config import get_service_provider_config
 
 load_dotenv()
 # initial service setup
 px.launch_app()
-llama_index.set_global_handler("arize_phoenix")
+llama_index.core.set_global_handler("arize_phoenix")
 # llama_index.set_global_handler("wandb", run_args={"project": "llamaindex"})
 openai.api_key = os.getenv("OPENAI_API_KEY")
 
@@ -32,16 +36,14 @@ IS_LOAD_FROM_VECTOR_STORE = True
 VDB_COLLECTION_NAME = "demo-v6"
 MODEL_NAME = ChatbotVersion.CHATGPT_4.value
 
-
 CHUNK_SIZE = 8191
 LLM, EMBED_MODEL = get_service_provider_config(
     service_provider=ServiceProvider.OPENAI, model_name=MODEL_NAME)
-service_context = ServiceContext.from_defaults(
-    chunk_size=CHUNK_SIZE,
-    llm=LLM,
-    embed_model=EMBED_MODEL,
-)
-set_global_service_context(service_context)
+
+LLM = Ollama(model="llama3.1:latest", request_timeout=60.0, context_window=10000)
+
+Settings.embed_model = EMBED_MODEL
+Settings.llm = LLM
 
 
 class AwesumIndexBuilder(IndexBuilder):
@@ -108,13 +110,13 @@ class AwesumCareToolChatbot(Chatbot):
             text_qa_template=CHAT_TEXT_QA_PROMPT)
 
     def _setup_tools(self):
-        from llama_index.tools.query_engine import QueryEngineTool
+        from llama_index.core.tools import QueryEngineTool
         self.tools = QueryEngineTool.from_defaults(
             query_engine=self.query_engine)
         return super()._setup_tools()
 
     def _setup_chat_engine(self):
-        from llama_index.agent import OpenAIAgent
+        from llama_index.agent.openai import OpenAIAgent
         self.chat_engine = OpenAIAgent.from_tools(
             tools=[self.tools],
             llm=LLM,
@@ -151,22 +153,9 @@ index_builder = AwesumIndexBuilder(vdb_collection_name=VDB_COLLECTION_NAME,
                                    is_load_from_vector_store=IS_LOAD_FROM_VECTOR_STORE)
 
 # gpt-3.5-turbo-1106, gpt-4-1106-preview
-awesum_chatbot = AwesumCareToolChatbot(model_name=model_name, index_builder=index_builder)
+# awesum_chatbot = AwesumCareToolChatbot(model_name=model_name, index_builder=index_builder)
 awesum_chatbot_context = AweSumCareContextChatbot(model_name=model_name, index_builder=index_builder)
-awesum_chatbot_simple = AweSumCareSimpleChatbot(model_name=model_name, index_builder=index_builder)
-
-
-def service_setup(model_name):
-    CHUNK_SIZE = 1024
-    LLM, EMBED_MODEL = get_service_provider_config(
-        service_provider=ServiceProvider.OPENAI, model_name=model_name)
-    service_context = ServiceContext.from_defaults(
-        chunk_size=CHUNK_SIZE,
-        llm=LLM,
-        embed_model=EMBED_MODEL,
-    )
-    set_global_service_context(service_context)
-    return LLM, EMBED_MODEL
+# awesum_chatbot_simple = AweSumCareSimpleChatbot(model_name=model_name, index_builder=index_builder)
 
 chatbot = gr.Chatbot(height=500)
 
@@ -184,7 +173,7 @@ with gr.Blocks(fill_height=True) as demo:
     with gr.Tab("With relevant context sent to system prompt"):
         context_interface = gr.ChatInterface(
             awesum_chatbot_context.stream_chat,
-            examples=awesum_chatbot.CHAT_EXAMPLES,
+            examples=AwesumCareToolChatbot.CHAT_EXAMPLES,
             chatbot=chatbot
         )
 

chat_template.py

@@ -1,5 +1,5 @@
-from llama_index.llms.base import ChatMessage, MessageRole
-from llama_index.prompts.base import ChatPromptTemplate
+from llama_index.core.llms import ChatMessage, MessageRole
+from llama_index.core import ChatPromptTemplate
 
 # text qa prompt
 TEXT_QA_SYSTEM_PROMPT = ChatMessage(

chatbot.py

@@ -8,8 +8,7 @@ from dotenv import load_dotenv
 from openai import OpenAI
 import phoenix as px
 import llama_index
-from llama_index import OpenAIEmbedding
-from llama_index.llms import ChatMessage, MessageRole
+from llama_index.core.llms import ChatMessage, MessageRole
 
 load_dotenv()
 

custom_io.py

@@ -11,8 +11,8 @@ from pathlib import Path
 import re
 from typing import Any, Dict, List, Optional
 
-from llama_index.readers.base import BaseReader
-from llama_index.readers.schema.base import Document
+from llama_index.core.readers.base import BaseReader
+from llama_index.core import Document
 
 
 class UnstructuredReader(BaseReader):

pyproject.toml

@@ -0,0 +1,32 @@
+[tool.poetry]
+name = "awesumcare-demo"
+version = "0.1.0"
+description = ""
+authors = ["Ray Yip"]
+readme = "README.md"
+package-mode = false
+
+[tool.poetry.dependencies]
+python = ">=3.10,<3.13"
+gradio = "^4.29.0"
+llama-index = "^0.10.58"
+nltk = "^3.8.1"
+phoenix = "^0.9.1"
+openai = "^1.37.0"
+python-dotenv = "^1.0.1"
+qdrant-client = "^1.10.1"
+arize-phoenix = "^4.14.1"
+fastapi = ">=0.80"
+llama-index-vector-stores-qdrant = "^0.2.14"
+llama-index-agent-openai = "^0.2.9"
+llama-index-embeddings-azure-openai = "^0.1.11"
+llama-index-embeddings-openai = "^0.1.11"
+llama-index-llms-openai = "^0.1.27"
+llama-index-callbacks-arize-phoenix = "^0.1.6"
+llama-index-llms-ollama = "^0.2.0"
+
+
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"

service_provider_config.py

@@ -1,8 +1,8 @@
 from dotenv import load_dotenv
-from llama_index import OpenAIEmbedding
-from llama_index.llms import OpenAI
-from llama_index.llms import AzureOpenAI
-from llama_index.embeddings import AzureOpenAIEmbedding
+from llama_index.embeddings.openai import OpenAIEmbedding
+from llama_index.llms.openai import OpenAI
+from llama_index.embeddings.azure_openai import AzureOpenAI
+from llama_index.embeddings.azure_openai import AzureOpenAIEmbedding
 from schemas import ServiceProvider, ChatbotVersion
 
 load_dotenv()