Merge pull request #22 from linjungz/staging

Add Support for OpenAI API
linjungz · Jul 9, 2023 · 4fb57cf · 4fb57cf
2 parents dd5d866 + 958b853
commit 4fb57cf
Show file tree

Hide file tree

Showing 2 changed files with 98 additions and 42 deletions.
diff --git a/README.md b/README.md
@@ -1,20 +1,21 @@
 # chat-with-your-doc
 
-`chat-with-your-doc` is a demonstration application that leverages the capabilities of Azure OpenAI ChatGPT/GPT-4 and LangChain to enable users to chat with their documents. This repository hosts the codebase, instructions, and resources needed to set up and run the application.
+`chat-with-your-doc` is a demonstration application that leverages the capabilities of ChatGPT/GPT-4 and LangChain to enable users to chat with their documents. This repository hosts the codebase, instructions, and resources needed to set up and run the application.
 
 ## Introduction
 
-The primary goal of this project is to simplify the interaction with documents and extract valuable information with using natural language. This project is built using LangChain and Azure OpenAI GPT-4/ChatGPT to deliver a smooth and natural conversational experience to the user.
+The primary goal of this project is to simplify the interaction with documents and extract valuable information with using natural language. This project is built using LangChain and GPT-4/ChatGPT to deliver a smooth and natural conversational experience to the user, with support for both `Azure OpenAI Services` and `OpenAI`
 
 ![](static/web_ui.png)
 
 ## Updates
 
-- 20230703 [v0.3.0](https://github.com/linjungz/chat-with-your-doc/releases/tag/v0.3.0): Web UI changed to Streamlit, with support for streaming
+- 20230709: Add Support for OpenAI API
+- 20230703: Web UI changed to Streamlit, with support for streaming
 
 ## Features
 
-- Upload documents as external knowledge base for Azure OpenAI GPT-4/ChatGPT.
+- Upload documents as external knowledge base for GPT-4/ChatGPT, support both `Azure OpenAI Services` and `OpenAI`
 - Support various format including PDF, DOCX, PPTX, TXT and etc.
 - Chat with the document content, ask questions, and get relevant answers based on the context.
 - User-friendly interface to ensure seamless interaction.
@@ -63,9 +64,11 @@ pip install -r requirements.txt
 
 ## Configuration
 
+### Azure OpenAI Services
+
 1. Obtain your Azure OpenAI API key, Endpoint and Deployment Name from the [Azure Portal](https://portal.azure.com/).
 
-2. Create `.env` in the root dir and set the environment variable in the file:
+2. Create `.env` in the root dir and set the environment variables in the file:
 
 ```
 OPENAI_API_BASE=https://your-endpoint.openai.azure.com
@@ -76,6 +79,16 @@ OPENAI_EMBEDDING_DEPLOYMENT_NAME=your-embedding-deployment-name
 Here's where you can find the deployment names for GPT and Embedding:
 ![Alt text](./static/deployment.png)
 
+### OpenAI
+
+1. Obtain your OpenAI API key from the [platform.openai.com](https://platform.openai.com/account/api-keys).
+
+2. Create `.env` in the root dir and set the environment variable in the file:
+
+```
+OPENAI_API_KEY=your-key-here
+```
+
 ## Usage: Web
 
 This will initialize the application based on `Streamlit` and open up the user interface in your default web browser. You can now upload a document to create a knowledge base and start a conversation with it.

diff --git a/chatbot.py b/chatbot.py
@@ -2,6 +2,7 @@
 import openai
 from dotenv import load_dotenv
 from langchain.chat_models import AzureChatOpenAI
+from langchain.chat_models import ChatOpenAI
 from langchain.embeddings import OpenAIEmbeddings
 from langchain.callbacks.base import BaseCallbackHandler
 
@@ -18,6 +19,8 @@
 from typing import List
 import streamlit
 
+REQUEST_TIMEOUT = 10
+
 class StreamHandler(BaseCallbackHandler):
     def __init__(self, container, initial_text=""):
         self.container = container
@@ -29,57 +32,98 @@ def on_llm_new_token(self, token: str, **kwargs) -> None:
 
 
 class DocChatbot:
-    llm: AzureChatOpenAI
-    condens_question_llm: AzureChatOpenAI
+    llm: ChatOpenAI
+    condens_question_llm: ChatOpenAI
     embeddings: OpenAIEmbeddings
     vector_db: FAISS
     chatchain: BaseConversationalRetrievalChain
 
     def __init__(self) -> None:
-        #init for OpenAI GPT-4 and Embeddings
+        #init for LLM and Embeddings
         load_dotenv()
+        assert(os.getenv("OPENAI_API_KEY") is not None)
+        api_key = str(os.getenv("OPENAI_API_KEY"))
+        embedding_deployment = "text-embedding-ada-002"
+
+        #check if user is using API from openai.com or Azure OpenAI Service by inspecting the api key
+        if api_key.startswith("sk-"):
+            # user is using API from openai.com
+            assert(len(api_key) == 51)
+
+            self.llm = ChatOpenAI(
+                temperature=0,
+                openai_api_key=api_key,
+                request_timeout=REQUEST_TIMEOUT,
+            ) # type: ignore
+        else:
+            # user is using Azure OpenAI Service
+            assert(os.getenv("OPENAI_GPT_DEPLOYMENT_NAME") is not None)
+            assert(os.getenv("OPENAI_API_BASE") is not None)
+            assert(len(api_key) == 32)
+
+            self.llm = AzureChatOpenAI(
+                deployment_name=os.getenv("OPENAI_GPT_DEPLOYMENT_NAME"),
+                temperature=0,
+                openai_api_version="2023-05-15",
+                openai_api_type="azure",
+                openai_api_base=os.getenv("OPENAI_API_BASE"),
+                openai_api_key=api_key,
+                request_timeout=REQUEST_TIMEOUT,
+            ) # type: ignore
 
-        self.llm = AzureChatOpenAI(
-            deployment_name=os.getenv("OPENAI_GPT_DEPLOYMENT_NAME"),
-            temperature=0,
-            openai_api_version="2023-05-15",
-            openai_api_type="azure",
-            openai_api_base=os.getenv("OPENAI_API_BASE"),
-            openai_api_key=os.getenv("OPENAI_API_KEY"),
-            request_timeout=30,
-        ) # type: ignore
+            embedding_deployment = os.getenv("OPENAI_EMBEDDING_DEPLOYMENT_NAME")
 
         self.condens_question_llm = self.llm
 
         self.embeddings = OpenAIEmbeddings(
-            deployment=os.getenv("OPENAI_EMBEDDING_DEPLOYMENT_NAME"), 
+            deployment=embedding_deployment, 
             chunk_size=1
             ) # type: ignore
 
     def init_streaming(self, condense_question_container, answer_container) -> None:
-        self.llm = AzureChatOpenAI(
-            deployment_name=os.getenv("OPENAI_GPT_DEPLOYMENT_NAME"),
-            temperature=0,
-            openai_api_version="2023-05-15",
-            openai_api_type="azure",
-            openai_api_base=os.getenv("OPENAI_API_BASE"),
-            openai_api_key=os.getenv("OPENAI_API_KEY"),
-            request_timeout=30,
-            streaming=True,
-            callbacks=[StreamHandler(answer_container)]
-        ) # type: ignore
-
-        self.condens_question_llm = AzureChatOpenAI(
-            deployment_name=os.getenv("OPENAI_GPT_DEPLOYMENT_NAME"),
-            temperature=0,
-            openai_api_version="2023-05-15",
-            openai_api_type="azure",
-            openai_api_base=os.getenv("OPENAI_API_BASE"),
-            openai_api_key=os.getenv("OPENAI_API_KEY"),
-            request_timeout=30,
-            streaming=True,
-            callbacks=[StreamHandler(condense_question_container, "🤔...")]
-        ) # type: ignore
+        api_key = str(os.getenv("OPENAI_API_KEY"))
+        if api_key.startswith("sk-"):
+            # user is using API from openai.com
+            self.llm = ChatOpenAI(
+                temperature=0,
+                openai_api_key=api_key,
+                request_timeout=REQUEST_TIMEOUT,
+                streaming=True,
+                callbacks=[StreamHandler(answer_container)]
+            ) # type: ignore
+
+            self.condens_question_llm = ChatOpenAI(
+                temperature=0,
+                openai_api_key=api_key,
+                request_timeout=REQUEST_TIMEOUT,
+                streaming=True,
+                callbacks=[StreamHandler(condense_question_container, "🤔...")]
+            ) # type: ignore
+        else:
+            # user is using Azure OpenAI Service
+            self.llm = AzureChatOpenAI(
+                deployment_name=os.getenv("OPENAI_GPT_DEPLOYMENT_NAME"),
+                temperature=0,
+                openai_api_version="2023-05-15",
+                openai_api_type="azure",
+                openai_api_base=os.getenv("OPENAI_API_BASE"),
+                openai_api_key=os.getenv("OPENAI_API_KEY"),
+                request_timeout=REQUEST_TIMEOUT,
+                streaming=True,
+                callbacks=[StreamHandler(answer_container)]
+            ) # type: ignore
+
+            self.condens_question_llm = AzureChatOpenAI(
+                deployment_name=os.getenv("OPENAI_GPT_DEPLOYMENT_NAME"),
+                temperature=0,
+                openai_api_version="2023-05-15",
+                openai_api_type="azure",
+                openai_api_base=os.getenv("OPENAI_API_BASE"),
+                openai_api_key=os.getenv("OPENAI_API_KEY"),
+                request_timeout=REQUEST_TIMEOUT,
+                streaming=True,
+                callbacks=[StreamHandler(condense_question_container, "🤔...")]
+            ) # type: ignore
 
     def init_chatchain(self, chain_type : str = "stuff") -> None:
         # init for ConversationalRetrievalChain
@@ -180,7 +224,6 @@ def init_vector_db_from_documents(self, file_list: List[str]):
 
         print("Generating embeddings and ingesting to vector db.")
         self.vector_db = FAISS.from_documents(docs, self.embeddings)
-        print(self.vector_db)
         print("Vector db initialized.")