Using LlamaIndex (GPT Index) with Azure OpenAI Service

gptindex_with_azure_openai_service.py

import os
import openai
from dotenv import load_dotenv
from llama_index import GPTSimpleVectorIndex, SimpleDirectoryReader, LLMPredictor, PromptHelper
from langchain.llms import AzureOpenAI
from langchain.embeddings import OpenAIEmbeddings
from llama_index import LangchainEmbedding

# Load env variables (create .env with OPENAI_API_KEY and OPENAI_API_BASE)
load_dotenv()

# Configure OpenAI API
openai.api_type = "azure"
openai.api_version = "2022-12-01"
openai.api_base = os.getenv('OPENAI_API_BASE')
openai.api_key = os.getenv("OPENAI_API_KEY")

deployment_name = "text-davinci-003"

# Create LLM via Azure OpenAI Service
llm = AzureOpenAI(deployment_name=deployment_name)
llm_predictor = LLMPredictor(llm=llm)
embedding_llm = LangchainEmbedding(OpenAIEmbeddings())

# Define prompt helper
max_input_size = 3000
num_output = 256
chunk_size_limit = 1000 # token window size per document
max_chunk_overlap = 20 # overlap for each token fragment
prompt_helper = PromptHelper(max_input_size=max_input_size, num_output=num_output, max_chunk_overlap=max_chunk_overlap, chunk_size_limit=chunk_size_limit)

# Read txt files from data directory
documents = SimpleDirectoryReader('data').load_data()
index = GPTSimpleVectorIndex(documents, llm_predictor=llm_predictor, embed_model=embedding_llm, prompt_helper=prompt_helper)
index.save_to_disk("index.json")

# Query index with a question
response = index.query("What is azure openai service?")
print(response)

I did the following.

The generate function called a few different llama hub loader I was using and created static files.

` llm = ChatOpenAI(deployment_id="gpt-4-32k-0314")
llm_predictor = LLMPredictor(llm=llm)
embedding_llm = LangchainEmbedding(OpenAIEmbeddings())

options = parse_args()
if options.generate:
    generate()

#load PDF from Confelunce
documents = SimpleDirectoryReader(DATA_DIR).load_data()

# Define prompt helper
max_input_size = 3000
num_output = 256
chunk_size_limit = 1000
max_chunk_overlap = 20
prompt_helper = PromptHelper(max_input_size=max_input_size, num_output=num_output, max_chunk_overlap=max_chunk_overlap, chunk_size_limit=chunk_size_limit)

service_context = ServiceContext.from_defaults(
   llm_predictor=llm_predictor, prompt_helper=prompt_helper, embed_model=embedding_llm
   )
# Create index
index = GPTSimpleVectorIndex.from_documents(documents, service_context=service_context)
index.save_to_disk("index.json")`

https://gist.github.com/csiebler/e287b791333011183792c08bad1dc140?permalink_comment_id=4558026#gistcomment-4558026

@csiebler
Did not work for me.
With the following Packages:

openai==0.27.6
llama-index==0.6.0
langchain==0.0.157
python-dotenv

I still got the API Error:
openai.error.InvalidRequestError: The API deployment for this resource does not exist. If you created the deployment within the last 5 minutes, please wait a moment and try again.

I have in Azure a Deplyoment called: gpt-35-turbo

I run this part of Code, was working:
https://gist.github.com/csiebler/e287b791333011183792c08bad1dc140?permalink_comment_id=4546610#gistcomment-4546610

Any Ideas?

https://gist.github.com/csiebler/e287b791333011183792c08bad1dc140?permalink_comment_id=4558026#gistcomment-4558026

@csiebler Did not work for me. With the following Packages:

openai==0.27.6
llama-index==0.6.0
langchain==0.0.157
python-dotenv

I still got the API Error: openai.error.InvalidRequestError: The API deployment for this resource does not exist. If you created the deployment within the last 5 minutes, please wait a moment and try again.

I have in Azure a Deplyoment called: gpt-35-turbo

I run this part of Code, was working: https://gist.github.com/csiebler/e287b791333011183792c08bad1dc140?permalink_comment_id=4546610#gistcomment-4546610

Any Ideas?

.....
I had deploy as well the training Model....

👀 prompt wizards, come share the best prompts with God Tier Prompts ✨