Remove Davinci from Workshop

akshata29 · Aug 2, 2023 · 46465f4 · 46465f4
1 parent ece444f
commit 46465f4
Show file tree

Hide file tree

Showing 21 changed files with 1,249 additions and 1,060 deletions.
diff --git a/Workshop/0A_PromptEngg.ipynb b/Workshop/0A_PromptEngg.ipynb
diff --git a/Workshop/0B_EmbeddingCluster.ipynb b/Workshop/0B_EmbeddingCluster.ipynb
diff --git a/Workshop/0_SetupandConfig.ipynb b/Workshop/0_SetupandConfig.ipynb
diff --git a/Workshop/10_PibCoPilot.ipynb b/Workshop/10_PibCoPilot.ipynb
@@ -25,7 +25,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 245,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -46,13 +46,12 @@
     "openAiEndPoint = f\"https://{OpenAiService}.openai.azure.com\"\n",
     "assert openAiEndPoint, \"ERROR: Azure OpenAI Endpoint is missing\"\n",
     "assert \"openai.azure.com\" in openAiEndPoint.lower(), \"ERROR: Azure OpenAI Endpoint should be in the form: \\n\\n\\t<your unique endpoint identifier>.openai.azure.com\"\n",
-    "openai.api_base = openAiEndPoint\n",
-    "davincimodel = OpenAiDavinci\n"
+    "openai.api_base = openAiEndPoint"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 246,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -69,7 +68,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 247,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -100,7 +99,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 248,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -110,15 +109,8 @@
     "    openai.api_key = OpenAiKey\n",
     "    openai.api_version = OpenAiVersion\n",
     "    openai.api_base = OpenAiBase\n",
-    "\n",
-    "    llm = AzureOpenAI(deployment_name=OpenAiDavinci,\n",
-    "            temperature=temperature,\n",
-    "            openai_api_key=OpenAiKey,\n",
-    "            max_tokens=tokenLength,\n",
-    "            batch_size=10, \n",
-    "            max_retries=12)\n",
     "    \n",
-    "    llmChat = AzureChatOpenAI(\n",
+    "    llm = AzureChatOpenAI(\n",
     "                openai_api_base=openai.api_base,\n",
     "                openai_api_version=OpenAiVersion,\n",
     "                deployment_name=OpenAiChat16k,\n",
@@ -136,20 +128,14 @@
     "    openai.api_key = OpenAiApiKey\n",
     "    llm = OpenAI(temperature=temperature,\n",
     "            openai_api_key=OpenAiApiKey,\n",
+    "            model_name=\"gpt-3.5-turbo\",\n",
     "            max_tokens=tokenLength)\n",
-    "    embeddings = OpenAIEmbeddings(openai_api_key=OpenAiApiKey)\n",
-    "\n",
-    "    llmChat = ChatOpenAI(temperature=temperature,\n",
-    "        openai_api_key=OpenAiApiKey,\n",
-    "        model_name=\"gpt-3.5-turbo\",\n",
-    "        max_tokens=tokenLength)\n",
-    "    \n",
-    "    embeddings = OpenAIEmbeddings(openai_api_key=OpenAiApiKey)"
+    "    embeddings = OpenAIEmbeddings(openai_api_key=OpenAiApiKey)    \n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 249,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -164,7 +150,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 250,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -174,7 +160,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 251,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -191,7 +177,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 252,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -200,7 +186,7 @@
        "'73124'"
       ]
      },
-     "execution_count": 252,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -211,12 +197,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 254,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
-    "# deletePibData(SearchService, SearchKey, pibIndexName, \"73124\", \"4\", returnFields=['id', 'symbol', 'cik', 'step', 'description', 'insertedDate',\n",
-    "#                                                                       'pibData'])"
+    "# deletePibData(SearchService, SearchKey, pibIndexName, \"1535929\", \"2\", returnFields=['id', 'symbol', 'cik', 'step', 'description', 'insertedDate',\n",
+    "#                                                                        'pibData'])"
    ]
   },
   {
@@ -288,18 +274,11 @@
     "            \"\"\"\n",
     "        optimizedPrompt = qaPromptTemplate.format(query=query)\n",
     "        qaPrompt = PromptTemplate(input_variables=[\"query\"],template=qaPromptTemplate)\n",
-    "        chain = LLMChain(llm=llmChat, prompt=qaPrompt)\n",
+    "        chain = LLMChain(llm=llm, prompt=qaPrompt)\n",
     "        q = chain.run(query=query)\n",
-    "        # completion = openai.Completion.create(\n",
-    "        #             engine=OpenAiDavinci,\n",
-    "        #             prompt=optimizedPrompt,\n",
-    "        #             temperature=temperature,\n",
-    "        #             max_tokens=100,\n",
-    "        #             n=1)\n",
-    "        # q = completion.choices[0].text\n",
     "        bingSearch = BingSearchAPIWrapper(k=20)\n",
     "        results = bingSearch.run(query=q)\n",
-    "        chain = load_summarize_chain(llmChat, chain_type=\"stuff\")\n",
+    "        chain = load_summarize_chain(llm, chain_type=\"stuff\")\n",
     "        docs = [Document(page_content=results)]\n",
     "        summary = chain.run(docs)\n",
     "        step1Executives.append({\n",
@@ -626,7 +605,7 @@
     "    ]\n",
     "\n",
     "    for question in commonQuestions:\n",
-    "        answer = findAnswer('stuff', 3, symbol, quarter, year, question, earningVectorIndexName, llmChat)\n",
+    "        answer = findAnswer('stuff', 3, symbol, quarter, year, question, earningVectorIndexName, llm)\n",
     "        if \"I don't know\" not in answer:\n",
     "            earningCallQa.append({\"question\": question, \"answer\": answer})\n",
     "\n",
@@ -642,7 +621,7 @@
     "        ]\n",
     "\n",
     "    for question in commonQuestions:\n",
-    "        answer = findAnswer('stuff', 3, symbol, quarter, year, question, earningVectorIndexName, llmChat)\n",
+    "        answer = findAnswer('stuff', 3, symbol, quarter, year, question, earningVectorIndexName, llm)\n",
     "        if \"I don't know\" not in answer:\n",
     "            earningCallQa.append({\"question\": question, \"answer\": answer})\n",
     "\n",
@@ -660,7 +639,7 @@
     "            \"\"\"\n",
     "    customPrompt = PromptTemplate(template=promptTemplate, input_variables=[\"text\"])\n",
     "    chainType = \"map_reduce\"\n",
-    "    summaryChain = load_summarize_chain(llmChat, chain_type=chainType, return_intermediate_steps=False, \n",
+    "    summaryChain = load_summarize_chain(llm, chain_type=chainType, return_intermediate_steps=False, \n",
     "                                combine_prompt=customPrompt)\n",
     "    summaryOutput = summaryChain({\"input_documents\": docs}, return_only_outputs=True)\n",
     "\n",
@@ -697,12 +676,12 @@
     "    customPrompt = PromptTemplate(template=promptTemplate, input_variables=[\"text\"])\n",
     "    try:\n",
     "        chainType = \"stuff\"\n",
-    "        summaryChain = load_summarize_chain(llmChat, chain_type=chainType, prompt=customPrompt)\n",
+    "        summaryChain = load_summarize_chain(llm, chain_type=chainType, prompt=customPrompt)\n",
     "        summaryOutput = summaryChain({\"input_documents\": docs}, return_only_outputs=True)\n",
     "        output = summaryOutput['output_text']\n",
     "    except:\n",
     "        chainType = \"map_reduce\"\n",
-    "        summaryChain = load_summarize_chain(llmChat, chain_type=chainType, combine_prompt=customPrompt)\n",
+    "        summaryChain = load_summarize_chain(llm, chain_type=chainType, combine_prompt=customPrompt)\n",
     "        summaryOutput = summaryChain({\"input_documents\": docs}, return_only_outputs=True)\n",
     "        output = summaryOutput['output_text']\n",
     "    s2Data.append({\n",
@@ -955,7 +934,7 @@
     "    i = 0\n",
     "    for pDocs in pressReleasesDocs:\n",
     "        try:\n",
-    "            outputAnswer = summarizePressReleases(llmChat, [pDocs])\n",
+    "            outputAnswer = summarizePressReleases(llm, [pDocs])\n",
     "            jsonStep = json.loads(outputAnswer)\n",
     "            pressReleasesPib.append({\n",
     "                    \"releaseDate\": last25PressReleases[i]['releaseDate'],\n",
@@ -1331,7 +1310,7 @@
     }
    ],
    "source": [
-    "print(getItem8Answer(llmChat, itemDocs8, \"What was the reported revenue for 2021?\", \"map_reduce\"))"
+    "print(getItem8Answer(llm, itemDocs8, \"What was the reported revenue for 2021?\", \"map_reduce\"))"
    ]
   },
   {
@@ -2147,7 +2126,7 @@
     "from langchain.tools.python.tool import PythonREPLTool\n",
     "from langchain.python import PythonREPL\n",
     "\n",
-    "agent = create_pandas_dataframe_agent(llmChat,df=incomeStmtDf, verbose=True)\n",
+    "agent = create_pandas_dataframe_agent(llm,df=incomeStmtDf, verbose=True)\n",
     "\n",
     "Question = 'What is the total revenue for the year 2022?'\n",
     "#Question = 'How much did revenue increased in 2022 in comparision to 2021?'\n",

diff --git a/Workshop/11_Prospectus.ipynb b/Workshop/11_Prospectus.ipynb
@@ -21,8 +21,7 @@
     "openAiEndPoint = f\"https://{OpenAiService}.openai.azure.com\"\n",
     "assert openAiEndPoint, \"ERROR: Azure OpenAI Endpoint is missing\"\n",
     "assert \"openai.azure.com\" in openAiEndPoint.lower(), \"ERROR: Azure OpenAI Endpoint should be in the form: \\n\\n\\t<your unique endpoint identifier>.openai.azure.com\"\n",
-    "openai.api_base = openAiEndPoint\n",
-    "davincimodel = OpenAiDavinci\n"
+    "openai.api_base = openAiEndPoint"
    ]
   },
   {

diff --git a/Workshop/1A_GenerateQuestions.ipynb b/Workshop/1A_GenerateQuestions.ipynb
@@ -39,8 +39,7 @@
     "openAiEndPoint = f\"https://{OpenAiService}.openai.azure.com\"\n",
     "assert openAiEndPoint, \"ERROR: Azure OpenAI Endpoint is missing\"\n",
     "assert \"openai.azure.com\" in openAiEndPoint.lower(), \"ERROR: Azure OpenAI Endpoint should be in the form: \\n\\n\\t<your unique endpoint identifier>.openai.azure.com\"\n",
-    "openai.api_base = openAiEndPoint\n",
-    "davincimodel = OpenAiDavinci\n"
+    "openai.api_base = openAiEndPoint"
    ]
   },
   {
@@ -53,7 +52,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -71,7 +70,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -116,7 +115,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -136,7 +135,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -145,7 +144,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -154,17 +153,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "[{'question': 'What are the steps to certify an item in Power BI?',\n",
-       "  'answer': \"To certify an item in Power BI, you need to get write permissions on the item, review it to ensure it meets your organization's certification standards, open the settings of the item, expand the endorsement section and select Certified, and finally select Apply.\"}]"
+       "[{'question': 'What are the steps to certify an item?',\n",
+       "  'answer': '1. Get write permissions on the item you want to certify. 2. Review the item and determine if it meets certification standards. 3. Open the settings of the item and select Certified. 4. Select Apply.'}]"
       ]
      },
-     "execution_count": 22,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }

diff --git a/Workshop/1_LoadData.ipynb b/Workshop/1_LoadData.ipynb
@@ -31,21 +31,21 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
-    "%pip install ./azure_search_documents-11.4.0b4-py3-none-any.whl"
+    "#%pip install ./azure_search_documents-11.4.0b4-py3-none-any.whl"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [],
    "source": [
     "# Install langchain\n",
-    "%pip install langchain"
+    "#%pip install langchain"
    ]
   },
   {
@@ -58,7 +58,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -79,8 +79,7 @@
     "openAiEndPoint = f\"https://{OpenAiService}.openai.azure.com\"\n",
     "assert openAiEndPoint, \"ERROR: Azure OpenAI Endpoint is missing\"\n",
     "assert \"openai.azure.com\" in openAiEndPoint.lower(), \"ERROR: Azure OpenAI Endpoint should be in the form: \\n\\n\\t<your unique endpoint identifier>.openai.azure.com\"\n",
-    "openai.api_base = openAiEndPoint\n",
-    "davincimodel = OpenAiDavinci\n"
+    "openai.api_base = openAiEndPoint"
    ]
   },
   {
@@ -93,7 +92,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -118,12 +117,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
     "# Flexibility to change the call to OpenAI or Azure OpenAI\n",
-    "embeddingModelType = \"openai\""
+    "embeddingModelType = \"azureopenai\""
    ]
   },
   {