Merge pull request #88 from wuerzelchen/main

refactor(04/04-streamlit): upgraded to use langchain 0.1.10

labs/04-deploy-ai/04-streamlit/app.py

@@ -2,61 +2,56 @@
 from langchain.callbacks.base import BaseCallbackHandler
 from langchain.chains import RetrievalQAWithSourcesChain
 from langchain.chains import RetrievalQA
-from langchain.chains import ConversationalRetrievalChain,LLMChain, RetrievalQA
+from langchain.chains import ConversationalRetrievalChain, LLMChain, RetrievalQA
 
 import os
 from dotenv import load_dotenv
 import qdrant_client
 
 load_dotenv()
 
-openai_api_type = os.getenv("OPENAI_API_TYPE")
-openai_api_key = os.getenv("OPENAI_API_KEY")
-openai_api_base = os.getenv("OPENAI_API_BASE")
 openai_api_version = os.getenv("OPENAI_API_VERSION")
 deployment_name = os.getenv("OPENAI_COMPLETION_DEPLOYMENT_NAME")
-completion_model = os.getenv("OPENAI_COMPLETION_MODEL")
 embedding_name = os.getenv("OPENAI_EMBEDDING_DEPLOYMENT_NAME")
 qdrant_url = os.getenv("QDRANT_URL")
 qdrant_collection = os.getenv("QDRANT_COLLECTION")
 
 st.set_page_config(page_title="Movie expert AI", page_icon="📺")
 
+
 def settings():
 
     # Vectorstore
     from langchain.vectorstores import Qdrant
-    from langchain.embeddings import OpenAIEmbeddings
-    embeddings_model = OpenAIEmbeddings(
-        deployment=embedding_name,
+    from langchain_openai import AzureOpenAIEmbeddings
+
+    embeddings_model = AzureOpenAIEmbeddings(
+        azure_deployment=embedding_name,
         chunk_size=1
-    ) 
+    )
 
     client = qdrant_client.QdrantClient(
         qdrant_url
     )
 
     vector_store = Qdrant(
-        client=client, collection_name=qdrant_collection, 
+        client=client, collection_name=qdrant_collection,
         embeddings=embeddings_model,
     )
 
     # LLM
-    from langchain.llms import AzureOpenAI
-    from langchain.chat_models import AzureChatOpenAI
+    from langchain_openai import AzureChatOpenAI
+
     llm = AzureChatOpenAI(
-        openai_api_type = openai_api_type,
-        openai_api_version = openai_api_version,
-        openai_api_base = openai_api_base,
-        openai_api_key = openai_api_key,
-        deployment_name = deployment_name,
-        model_name=completion_model
+        openai_api_version=openai_api_version,
+        deployment_name=deployment_name
     )
 
     retriever = vector_store.as_retriever(search_type="mmr")
 
     return retriever, llm
 
+
 class StreamHandler(BaseCallbackHandler):
     def __init__(self, container, initial_text=""):
         self.container = container
@@ -92,15 +87,17 @@ def on_retriever_end(self, documents, **kwargs):
 retriever = st.session_state.retriever
 llm = st.session_state.llm
 
-# User input 
+# User input
 question = st.text_input("`Ask a question:`")
 
 if question:
-    qa_sources_chain = RetrievalQAWithSourcesChain.from_chain_type(llm, retriever=retriever)
+    qa_sources_chain = RetrievalQAWithSourcesChain.from_chain_type(
+        llm, retriever=retriever)
     retrieval_streamer_cb = PrintRetrievalHandler(st.container())
     answer = st.empty()
     stream_handler = StreamHandler(answer, initial_text="`Answer:`\n\n")
-    result = qa_sources_chain({"question": question},callbacks=[retrieval_streamer_cb, stream_handler])
+    result = qa_sources_chain({"question": question}, callbacks=[
+                              retrieval_streamer_cb, stream_handler])
     print(result)
     answer.info('`Answer:`\n\n' + result['answer'])
-    st.info('`Sources:`\n\n' + result['sources'])
+    st.info('`Sources:`\n\n' + result['sources'])

labs/04-deploy-ai/04-streamlit/requirements.txt

@@ -1,6 +1,6 @@
 streamlit==1.28.2
-langchain==0.0.331
-qdrant-client==1.4.0
-openai==0.28.1
+langchain==0.1.10
+qdrant-client==1.7.3
 html2text==2020.1.16
-tiktoken==0.2.0
+tiktoken==0.6.0
+python-dotenv==1.0.1