Adding chat completion to qdrant and acs

Azure · Sep 14, 2023 · 73acfed · 73acfed
1 parent d89f1b4
commit 73acfed
Show file tree

Hide file tree

Showing 2 changed files with 16 additions and 8 deletions.
diff --git a/labs/03-orchestration/03-Qdrant/qdrant.ipynb b/labs/03-orchestration/03-Qdrant/qdrant.ipynb
@@ -75,15 +75,16 @@
    "source": [
     "from langchain.embeddings import OpenAIEmbeddings\n",
     "from langchain.llms import AzureOpenAI\n",
+    "from langchain.chat_models import AzureChatOpenAI\n",
     "\n",
     "# Create an Embeddings Instance of Azure OpenAI\n",
     "embeddings = OpenAIEmbeddings(\n",
     "    deployment=embedding_name,\n",
     "    chunk_size=1\n",
     ") \n",
     "\n",
-    "# Create a Completion Instance of Azure OpenAI\n",
-    "llm = AzureOpenAI(\n",
+    "# Create a Chat Completion Instance of Azure OpenAI\n",
+    "llm = AzureChatOpenAI(\n",
     "    openai_api_type = openai_api_type,\n",
     "    openai_api_version = openai_api_version,\n",
     "    openai_api_base = openai_api_base,\n",
@@ -227,7 +228,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "llm = AzureOpenAI(\n",
+    "llm = AzureChatOpenAI(\n",
     "    openai_api_type = openai_api_type,\n",
     "    openai_api_version = openai_api_version,\n",
     "    openai_api_base = openai_api_base,\n",
@@ -236,14 +237,20 @@
     "    model_name=\"gpt-35-turbo\"\n",
     ")\n",
     "chain = RetrievalQA.from_chain_type(llm=llm, chain_type=\"stuff\", retriever=docsearch.vectorstore.as_retriever(), input_key=\"question\", return_source_documents=True)\n",
+    "\n",
     "query = \"Do you have a column called popularity?\"\n",
+    "print (\"First query: \" + query)\n",
     "response = chain({\"question\": query})\n",
     "print(response['result'])\n",
+    "print (\"Source documents for first query\")\n",
     "print(response['source_documents'])\n",
+    "print (\"\\n\")\n",
     "\n",
     "query = \"What is the movie with the highest popularity?\"\n",
+    "print (\"Second query: \" + query)\n",
     "response = chain({\"question\": query})\n",
     "print(response['result'])\n",
+    "print(\"Source Documents for second query\")\n",
     "print(response['source_documents'])"
    ]
   },
@@ -324,7 +331,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.11.5"
   },
   "orig_nbformat": 4
  },

diff --git a/labs/03-orchestration/04-ACS/acs.ipynb b/labs/03-orchestration/04-ACS/acs.ipynb
@@ -5,7 +5,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# 03 - Langchain with Azure Cognitive Search\n",
+    "# 04 - Langchain with Azure Cognitive Search\n",
     "\n",
     "In this lab, we will do a deeper dive around the Azure Cognitive Search (ACS) vector store and different ways to interact with it."
    ]
@@ -29,7 +29,7 @@
    "source": [
     "RG=\"azure-cognitive-search-rg\"\n",
     "LOC=\"westeurope\"\n",
-    "NAME=\"acs-vectorstore-<INITIALS>\"\n",
+    "NAME=\"acs-vectorstore-mtjw\"\n",
     "!az group create --name $RG --location $LOC\n",
     "!az search service create -g $RG -n $NAME -l $LOC --sku Basic --partition-count 1 --replica-count 1"
    ]
@@ -122,6 +122,7 @@
    "source": [
     "from langchain.embeddings.openai import OpenAIEmbeddings\n",
     "from langchain.llms import AzureOpenAI\n",
+    "from langchain.chat_models import AzureChatOpenAI\n",
     "\n",
     "# Create an Embeddings Instance of Azure OpenAI\n",
     "embeddings = OpenAIEmbeddings(\n",
@@ -130,7 +131,7 @@
     ") \n",
     "\n",
     "# Create a Completion Instance of Azure OpenAI\n",
-    "llm = AzureOpenAI(\n",
+    "llm = AzureChatOpenAI(\n",
     "    openai_api_type = openai_api_type,\n",
     "    openai_api_version = openai_api_version,\n",
     "    openai_api_base = openai_api_base,\n",
@@ -322,7 +323,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.11.5"
   },
   "orig_nbformat": 4
  },