Tag Archives: microsoft
Microsoft Copilot in Azure Series – Data Collection and Feedback
Hello folks!
Today, we’re continuing our coverage of Microsoft Copilot in Azure. We’ll talk about how to provide feedback and because we’ve been getting this question multiple times, what data is collected by Microsoft Copilot in Azure.
So, buckle up and let’s get started!
First off, let’s talk about the data Copilot collects. You know those prompts you give and the responses you get? well, They’re never used to train or improve the Azure OpenAI Service foundation models. They’re only used to improve Microsoft products and services, and only when you click the give a thumbs up, or down and EXPLICITELY click the share radio button to share the prompt and generated responses.
On top of that, once you provide consent, we only log the prompt and responses for that conversation only. We do not collect other conversations for that user, or for that session. The consent is for just that single conversation.
All that info is stored and used as per the Microsoft Privacy Statement. You can find the full statement right here: https://aka.ms/CopilotInAzure/PrivacyStatement
Now. Going back to the feedback I mentioned a minute ago.
You don’t need to get mad if something does not go as expected. No need to ask for the manager.
We have a built-in way of providing feedback. It’s the way you can have your say about Microsoft Copilot in Azure.
Important
Microsoft Copilot in Azure (preview) is currently in PREVIEW. See the Supplemental Terms of Use for Microsoft Azure Previews for legal terms that apply to Azure features that are in beta, preview, or otherwise not yet released into general availability.
The video attached above shows you how to provide feedback on the results you get whether you are happy or not.
That’s it for today! If you’ve got any more questions or need some help, don’t hesitate to drop a comment in the section below.
And that’s a wrap on how to provide feedback and how Microsoft Copilot in Azure collects and uses data and stay tuned for more updates!
cheers!
Pierre Roman
Microsoft Tech Community – Latest Blogs –Read More
Teams Live: Why do some people joining as presenters/event group and not audience?
Hi! I was a producer in a Teams Live event earlier today. I shared the link with all employees using the “get attendee link” feature under the ‘Invite attendees’ header.
However, for some reason, around 200 people didn’t join the live event as audience members because they needed to be admitted. After that, their names appeared under the ‘event group’ tab. I also noticed that these 200 people could access the chat feature and unmute themselves, which means they joined as presenters. This was distracting (and beat the “live” purposes), especially because these people sometimes joined without muting themselves, and the lobby noise kept popping up (to all of these ‘presenters’ as well).
I tracked the link that I shared in all my communications, but it was the same link in all instances. I wonder why this happened? Why could some people join as regular attendees while others joined as presenters if they were given the same attendee link? Is there any way to prevent this from happening again?
Thank you in advance 🙂
Hi! I was a producer in a Teams Live event earlier today. I shared the link with all employees using the “get attendee link” feature under the ‘Invite attendees’ header. However, for some reason, around 200 people didn’t join the live event as audience members because they needed to be admitted. After that, their names appeared under the ‘event group’ tab. I also noticed that these 200 people could access the chat feature and unmute themselves, which means they joined as presenters. This was distracting (and beat the “live” purposes), especially because these people sometimes joined without muting themselves, and the lobby noise kept popping up (to all of these ‘presenters’ as well). I tracked the link that I shared in all my communications, but it was the same link in all instances. I wonder why this happened? Why could some people join as regular attendees while others joined as presenters if they were given the same attendee link? Is there any way to prevent this from happening again? Thank you in advance 🙂 Read More
How can I create Windows 11 bootable usb from ISO?
Hi everyone! I recently needed to make a bootable USB drive from a Windows 11 ISO file, but I ran into some problems. I tried using Rufus, but it didn’t recognize the USB drive on my computer. I’m not sure what the problem is, or what other tools or methods I can use. Can anyone recommend some reliable tools, or explain the steps in detail? Thank you very much for your help!
Hi everyone! I recently needed to make a bootable USB drive from a Windows 11 ISO file, but I ran into some problems. I tried using Rufus, but it didn’t recognize the USB drive on my computer. I’m not sure what the problem is, or what other tools or methods I can use. Can anyone recommend some reliable tools, or explain the steps in detail? Thank you very much for your help! Read More
Security Update for SQL Server 2022 RTM GDR
The Security Update for SQL Server 2022 RTM GDR is now available for download at the Microsoft Download Center and Microsoft Update Catalog sites. This package cumulatively includes all previous security fixes for SQL Server 2022 RTM, plus it includes the new security fixes detailed in the KB Article.
Security Bulletins:
Security Update of SQL Server 2022 RTM GDR KB Article: KB5040936
Microsoft Download Center: https://www.microsoft.com/download/details.aspx?id=054a7ae7-1535-43f7-bb5b-398b465ee16e
Microsoft Update Catalog: https://www.catalog.update.microsoft.com/Search.aspx?q=5040936
Latest Updates for Microsoft SQL Server: https://learn.microsoft.com/en-us/troubleshoot/sql/releases/download-and-install-latest-updates
Microsoft Tech Community – Latest Blogs –Read More
Security Update for SQL Server 2022 RTM CU13
The Security Update for SQL Server 2022 RTM CU13 is now available for download at the Microsoft Download Center and Microsoft Update Catalog sites. This package cumulatively includes all previous security fixes for SQL Server 2022 RTM CU13, plus it includes the new security fixes detailed in the KB Article.
Security Bulletins:
Security Update of SQL Server 2022 RTM CU13 KB Article: KB5040939
Microsoft Download Center: https://www.microsoft.com/download/details.aspx?familyid=f4666419-9abb-4b7e-889a-61e845cff8fe
Microsoft Update Catalog: https://www.catalog.update.microsoft.com/Search.aspx?q=5040939
Latest Updates for Microsoft SQL Server: https://learn.microsoft.com/en-us/troubleshoot/sql/releases/download-and-install-latest-updates
Microsoft Tech Community – Latest Blogs –Read More
Security Update for SQL Server 2019 RTM GDR
The Security Update for SQL Server 2019 RTM GDR is now available for download at the Microsoft Download Center and Microsoft Update Catalog sites. This package cumulatively includes all previous security fixes for SQL Server 2019 RTM, plus it includes the new security fixes detailed in the KB Article.
Security Bulletins:
Security Update of SQL Server 2019 RTM GDR KB Article: KB5040986
Microsoft Download Center: https://www.microsoft.com/download/details.aspx?id=349b6b1d-8f7b-4be8-9ed4-94a7ffc186a7
Microsoft Update Catalog: https://www.catalog.update.microsoft.com/Search.aspx?q=5040986
Latest Updates for Microsoft SQL Server: https://learn.microsoft.com/en-us/troubleshoot/sql/releases/download-and-install-latest-updates
Microsoft Tech Community – Latest Blogs –Read More
Security Update for SQL Server 2019 RTM CU27
The Security Update for SQL Server 2019 RTM CU27 is now available for download at the Microsoft Download Center and Microsoft Update Catalog sites. This package cumulatively includes all previous security fixes for SQL Server 2019 RTM CU27, plus it includes the new security fixes detailed in the KB Article.
Security Bulletins:
Security Update of SQL Server 2019 RTM CU27 KB Article: KB5040948
Microsoft Download Center: https://www.microsoft.com/download/details.aspx?id=7205bb55-2221-4367-9210-9030d83da05f
Microsoft Update Catalog: https://www.catalog.update.microsoft.com/Search.aspx?q=5040948
Latest Updates for Microsoft SQL Server: https://learn.microsoft.com/en-us/troubleshoot/sql/releases/download-and-install-latest-updates
Microsoft Tech Community – Latest Blogs –Read More
Security Update for SQL Server 2017 RTM GDR
The Security Update for SQL Server 2017 RTM GDR is now available for download at the Microsoft Download Center and Microsoft Update Catalog sites. This package cumulatively includes all previous security fixes for SQL Server 2017 RTM, plus it includes the new security fixes detailed in the KB Article.
Security Bulletins:
Security Update of SQL Server 2017 RTM GDR KB Article: KB5040942
Microsoft Download Center: https://www.microsoft.com/download/details.aspx?id=119a86c0-ad18-4f6f-b763-d1e994ccc5be
Microsoft Update Catalog: https://www.catalog.update.microsoft.com/Search.aspx?q=5040942
Latest Updates for Microsoft SQL Server: https://learn.microsoft.com/en-us/troubleshoot/sql/releases/download-and-install-latest-updates
Microsoft Tech Community – Latest Blogs –Read More
Security Update for SQL Server 2017 RTM CU31
The Security Update for SQL Server 2017 RTM CU31 is now available for download at the Microsoft Download Center and Microsoft Update Catalog sites. This package cumulatively includes all previous security fixes for SQL Server 2017 RTM CU31, plus it includes the new security fixes detailed in the KB Article.
Security Bulletins:
Security Update of SQL Server 2017 RTM CU31 KB Article: KB5040940
Microsoft Download Center: https://www.microsoft.com/download/details.aspx?id=324422a6-abc8-44ac-984a-8ab149013eff
Microsoft Update Catalog: https://www.catalog.update.microsoft.com/Search.aspx?q=5040940
Latest Updates for Microsoft SQL Server: https://learn.microsoft.com/en-us/troubleshoot/sql/releases/download-and-install-latest-updates
Microsoft Tech Community – Latest Blogs –Read More
Security Update for SQL Server 2016 SP3 GDR
The Security Update for SQL Server 2016 SP3 GDR is now available for download at the Microsoft Download Center and Microsoft Update Catalog sites. This package cumulatively includes all previous security fixes for SQL Server 2016 SP3, plus it includes the new security fixes detailed in the KB Article.
Security Bulletins:
Security Update of SQL Server 2016 SP3 GDR KB Article: KB5040946
Microsoft Download Center: https://www.microsoft.com/download/details.aspx?familyid=a17ab24d-f54f-440d-a38a-308a0bd7ac98
Microsoft Update Catalog: https://www.catalog.update.microsoft.com/Search.aspx?q=5040946
Latest Updates for Microsoft SQL Server: https://learn.microsoft.com/en-us/troubleshoot/sql/releases/download-and-install-latest-updates
Microsoft Tech Community – Latest Blogs –Read More
Security Update for SQL Server 2016 SP3 Azure Connect Feature Pack
The Security Update for SQL Server 2016 SP3 Azure Connect Feature Pack is now available for download at the Microsoft Download Center and Microsoft Update Catalog sites. This package cumulatively includes all previous SQL Server 2016 SP3 Azure Connect Feature Pack, plus it includes the new security fixes detailed in the KB Article.
Security Bulletins:
Security Update of SQL Server 2016 SP3 Azure Connect Feature Pack KB Article: KB5040944
Microsoft Download Center: https://www.microsoft.com/download/details.aspx?id=8a0eb2ca-1dec-47b0-a0b8-b87a3e34909f
Microsoft Update Catalog: https://www.catalog.update.microsoft.com/Search.aspx?q=5040944
Latest Updates for Microsoft SQL Server: https://learn.microsoft.com/en-us/troubleshoot/sql/releases/download-and-install-latest-updates
Microsoft Tech Community – Latest Blogs –Read More
Grid modernization in Excel
Hi, Microsoft 365 Insiders,
Great news for Excel enthusiasts! Microsoft Excel for the web has rolled out new grid functionality that will enhance your data management experience. Our latest updates enable effortless resizing, quicker insertion, smoother navigation and easier cell highlighting. Discover how this modernized functionality can streamline your workflow and help you get more done in less time. Check out our latest blog: Grid modernization in Excel
Thanks!
Perry Sjogren
Microsoft 365 Insider Community Manager
Become a Microsoft 365 Insider and gain exclusive access to new features and help shape the future of Microsoft 365. Join Now: Windows | Mac | iOS | Android
Hi, Microsoft 365 Insiders,
Great news for Excel enthusiasts! Microsoft Excel for the web has rolled out new grid functionality that will enhance your data management experience. Our latest updates enable effortless resizing, quicker insertion, smoother navigation and easier cell highlighting. Discover how this modernized functionality can streamline your workflow and help you get more done in less time. Check out our latest blog: Grid modernization in Excel
Thanks!
Perry Sjogren
Microsoft 365 Insider Community Manager
Become a Microsoft 365 Insider and gain exclusive access to new features and help shape the future of Microsoft 365. Join Now: Windows | Mac | iOS | Android Read More
Teams calling issues with Sales Force CRM System
Hello
Please i need your help on this issue .
Since updating to the new version of Teams users cannot make calls by clicking a phone number on the Sales Force CRM system.
Previously when a user clicked the phone number on the Sales Force system Teams would open and dial the number..
For example here is the Sale force CRM System. When they click on the phone number as shown:
They get the following error:
When they click on the phone number on the Sales Force CRM System it will redirect them to teams to call the user.
This happened when Teams updated to the new version. So i dont know if it is a problem with the New Teams.
Hello Please i need your help on this issue .Since updating to the new version of Teams users cannot make calls by clicking a phone number on the Sales Force CRM system. Previously when a user clicked the phone number on the Sales Force system Teams would open and dial the number.. For example here is the Sale force CRM System. When they click on the phone number as shown: They get the following error: When they click on the phone number on the Sales Force CRM System it will redirect them to teams to call the user. This happened when Teams updated to the new version. So i dont know if it is a problem with the New Teams. Read More
SharePoint List formatting a Date column
I have a SharePoint List. Two of the columns are “Due Date” and “Status”. The Due Date column is obviously, populate with dates, with times turned off. The Status field is a choice field, formatted in pills and contains Queued, Work in Progress, On-Hold, Completed and Cancelled. If the Date is prior to the current day, meaning past due, AND the status is NOT Cancelled or Completed, I want to format the Due Date text. I would like it to be Centered, Bold and Red. I would also like to make it a slightly larger text, if possible, using font size like 8, 10, 12 … so I can easily alter the code. I would like to keep the existing background color and also use hexadecimal code, #FF0000, to represent the red color, so I can easily tweak the color if needed.
I have a SharePoint List. Two of the columns are “Due Date” and “Status”. The Due Date column is obviously, populate with dates, with times turned off. The Status field is a choice field, formatted in pills and contains Queued, Work in Progress, On-Hold, Completed and Cancelled. If the Date is prior to the current day, meaning past due, AND the status is NOT Cancelled or Completed, I want to format the Due Date text. I would like it to be Centered, Bold and Red. I would also like to make it a slightly larger text, if possible, using font size like 8, 10, 12 … so I can easily alter the code. I would like to keep the existing background color and also use hexadecimal code, #FF0000, to represent the red color, so I can easily tweak the color if needed. Read More
Enhancing Document Extraction with Azure AI Document Intelligence and LangChain for RAG Workflows.
The broadening of conventional data engineering pipelines and applications to include document extraction and preprocessing for unstructured PDFs, audio, and video files is becoming more prevalent. This shift is propelled by the increasing demand for advanced generative AI applications in businesses, adhering to the RAG (Retrievable Augmented Generation) model. In this post, I will discuss a proof of concept that utilizes Azure AI Document Intelligence to augment these functionalities.
In a prior solution, I examined the capability of Azure AI Search to automatically vectorize data through its integrated vectorization feature. This feature manages the entire pipeline process, from ingestion and extraction to enrichment and uploading data to the search index, with minimal or no custom coding required. Nonetheless, I noted a limitation: the current skills, despite the option to import external ones, failed to extract all vital content from documents, such as embedded tables.
Introduction.
In this post and the accompanying notebook, I present a solution utilizing the prebuilt layout model from Azure AI Document Intelligence to extract essential content from a PDF document. Subsequently, it employs the semantic chunking feature rather than the fixed-length chunking option. This approach aims to address the limitations of the prior solution, enhancing the relevance and precision of search retrieval.
This is the first of two posts, which shows a solution that uses Azure AI Document Intelligence and LangChain to create a Retrieval Augmented Generation (RAG) workflow. It uses the LangChain Azure AI Document Intelligence document loader to ingest, extract and retrieve tables values, paragraphs, and layout information from a PDF file. The output is in markdown format, which is processed by LangChain’s markdown header splitter. This class supports the semantic chunking feature of Azure AI Document Intelligence service to produce semantic chunks of the source document.
We employ the Azure AI Search Python SDK to build the Azure AI Search index, load the semantically chunked documents into this index and execute a hybrid + semantic search query at the end of the notebook to assess the search result relevance.
Below is a straightforward architectural diagram of the solution:
Before this post extends too much and loses its appeal, let’s explore the essential elements of the solution architecture and the corresponding notebook code implementation that are pertinent to this discussion.
LangChain – Ingestion, Extraction and Semantic Chunking:
LangChain is a framework for developing LLM (large language model) powered applications. LangChain has built a huge collection of abstractions that enable the integration of LLMs to external data sources, user input and other components and services required to build GenAI application use cases. We introduced LangChain into this solution because it provided two key abstractions:
1. The Azure AI Document Intelligence Loader API offers an interface for loading data into Azure AI Document Intelligence and extracting the necessary content from documents. In this proof of concept, I am utilizing a PDF document that contains an embedded table. To facilitate the successful import of the PDF document from Azure Blob Storage, the RBAC role of Storage Blob Data Reader has been granted to the Document Intelligence managed identity for the Blob storage resource.
The ‘prebuilt layout’ extraction model for Document Intelligence is one of the parameter values provided during the ingestion task. It supports formatting of tables with column headers into key-value pairs (to enhance readability for the LLM), and each table row is transformed into a text line, while maintaining the original structure of the table values.
2. The MarkdownHeaderTextSplitter class divides the markdown file generated from an extracted PDF document according to specified headers. Since a markdown file is structured with headers, this organizes the content into semantic sections. The headers used for splitting are determined by the ‘headers_to_split_on‘ configuration.
Chunking is a crucial component in the development of any RAG-based solution. There are two primary methods of chunking: fixed-sized and semantic. This PoC aims to evaluate the relevance and accuracy of RAG responses using the semantic chunking method. This technique divides extracted content according to specific content headers or sections, as opposed to the fixed-sized method, which relies on an overlap configuration to preserve contextual relevance across text segments. Semantic chunking allows an application to identify semantically coherent fragments within sentences or paragraphs. These fragments can be processed separately and then reassembled into semantic representations without losing information, context, or semantic integrity. The text’s inherent meaning guides the chunking process. In any document extraction process, the chunking strategy requires careful consideration and planning, as it significantly impacts the relevance and accuracy of query responses in RAG-based solutions. Determining the size of tokens per page can aid in the implementation of chunking.
The text splitter’s output is a list of document objects, as demonstrated in the following code snippet from the notebook.:
Azure AI Search – Create and Load Vector Index:
Azure AI Search offers secure and rapid information retrieval at scale for data content within generative AI applications. In this scenario, the content of the PDF document, which is chunked, will be vectorized and stored in the search service to enable the millisecond retrieval times necessary for RAG-based applications. Such quick retrieval cannot be achieved using Azure Blob storage. The search service does not interact with the primary data store of blob storage, except when utilizing an Indexer.
I utilized the Azure AI Search Python SDK to construct the vector search index by first defining the schema, then initiating the data import process. For the vector search configuration, I chose the Hierarchical Navigable Small World (HNSW) algorithm over the Exhaustive k-nearest neighbors (KNN) because it’s more efficient. HNSW offers a scalable solution for nearest neighbor searches by quickly identifying approximate nearest neighbors, which is ideal for large-scale and high-dimensional data applications. Additional details on vector search configuration can be found here.
I am utilizing the Azure OpenAI text-embedding-ada-002 model to vectorize segmented content for vector search capabilities. An ID field for each document in the index is generated using a custom function that employs the base64 Python package. The code snippet below illustrates the configuration of the vector index and the data loading process for this proof of concept:
Azure AI Document Intelligence:
Azure AI Document Intelligence is a cloud-based Azure AI Service that enables the building of intelligent document processing solutions. In a previous exploration, I implemented the Azure AI Search integrated vectorization feature that enables an automated solution for data extraction, ingestion and retrieval for RAG applications. While it’s great for a lot of use cases, I wasn’t able to get answers to queries that required searching through embedded table data in the PDF file. Azure AI Document Intelligence has a wider range of document extraction models, with a lot more functionality for complex PDF documents and other file types. Two primary reasons for choosing this service:
The prebuilt layout model has the functionality to “crack” and extract data from most complex PDF documents. I directly ingested the PDF document from the Blob storage container.
The semantic chunking technique seems an exciting option that could potentially yield better answers and search responses than the fixed-sized chunking technique.
Semantic and Hybrid Search:
This search feature provided an accurate answer to my query regarding the health plan’s cost. The prior solution failed to address this question as the extraction skills applied, along with the underlying model, were unable to completely decipher the PDF document and extract the values from the embedded table.
Azure Functions – Semantic Kernel:
Azure Functions, a serverless computing service, hosts the semantic kernel and LangChain source code for the application. The semantic kernel, similar to LangChain, is an LLM framework that aids in the development of Generative AI applications. In a follow-up post, I will demonstrate how I utilized the Semantic Kernel to orchestrate the various components of the Generative AI application, integrating prompts, chat services, embeddings, and native functions (plugins), among others.
Conclusion:
This post showcases how to utilize Azure AI Document Intelligence to manually process a PDF document with its prebuilt layout model and extract all the content, including any embedded tables. The objective is to employ this content to ground the responses of an LLM within a knowledgebase Q&A agent.
Azure AI Document Intelligence is recommended for production use cases involving a variety of complex file types and formats like PDFs. It appears to be better suited to meet application requirements compared to the Azure AI Search integrated vectorization option. Additionally, I plan to explore the evaluation of generative AI applications using the prompt flow Python SDK, responsible prompt engineering tools and techniques, and content safety measures in Azure.
The code for the PoC is available in the following repository.
Microsoft Tech Community – Latest Blogs –Read More
Azure OpenAI Extension for Function Apps Hands-on Experience
This blog will give some insights on the newly released Azure OpenAI extension. It will combine both Azure OpenAI service and Azure Function Apps. We will discuss the following contents:
Why this extension?
What’s the current requirements and support scope?
How to use this?
Please note that due to fast-growing development of AI services, some contents may be outdated. This article will use the released version as of July 2024.
Why this extension?
Compared with standard Azure OpenAI API call, the extension would give:
Capability to work with large variety of triggers and bindings offered by function apps. Function Apps would have pre-defined triggers to allow developers control event-driven or routine-based tasks. Based on our tests, the extension would work well with the current offered types of functions.
Flexibility in the development phase when multiple Azure products are engaged. Different bindings allow function apps to listen and respond when certain Azure product changes. With host.json file inside Function Apps, settings would be easier to adjust and test.
What’s the current requirements and support scope?
To use this extension, the following requirements must be met:
Get access to Azure OpenAI service. If you are new to Azure, remember to request Azure OpenAI access via this link. After the approval, please go to deployment section and choose one of the LLMs (Large Language Models) for future use. Example:
Azure Function Apps under the following language version:
.NET 6+
Java 11+
Node.JS 18+
PowerShell 7.4
Python 3.9+
The extension will support all mentioned languages & version.
Support Scope: Public Preview. Since this feature is still in preview, you may notice some issues. If so, please reach out via support request, or raise issues via GitHub Issue Page pointing to the extension.
How to use this?
The following resources are highly recommended to start using this extension:
Extension Samples on GitHub
Official Microsoft Docs of this extension
The following demos will use .NET8 and GPT-4o as language / LLM. We will dive deeper into 3 sections:
Chat
Text Completion
Rag-cosmosDB
Essentially, this extension would help you make API calls to the Azure OpenAI endpoint with a smoother experience.
Chat
Chat allows users to communicate with Azure OpenAI service. The response would be generated based on pre-defined prompts and questions. Please do keep in mind if you want to have long-term memory of the chat history, 2 options are available when using the extension:
Customize the code to store all asked questions in the response body, then invoke API call to send all contents to Azure OpenAI Service endpoint. Downside is that the used token size would grow exponentially as you add more history.
Use Azure cosmosDB as the persistent storage, and leverage semantic search to find the desired chat history. This is a recommended solution, and we will cover later in the 3rd section.
[Local Debugging]
Clone this GitHub repo, you may take a look at README.md to see detailed instructions.
Go to csharp-ooproc/local.settings.json. Replace values to match parameters in your Azure OpenAI endpoint. Also, remember to add setting CHAT_MODEL_DEPLOYMENT_NAME = <name of your deployed LLM>
Head to ChatBot.cs file, you can review and learn how to invoke REST api call to the Azure OpenAI services. Modify the trigger type if needed.
Remember to install the NuGet package Microsoft.Azure.Functions.Worker.Extensions.OpenAI == 0.16.0 Alpha. You may add by executing .NET CLI in VSCode, or a simple click to install this package in Visual Studio.
Go to root folder → cd samples/chat/csharp-ooproc && func start. We will not demo how to test as this has already been covered in the GitHub repo. In all, we will leverage the below 3 API requests:
Functions:
CreateChatBot: [PUT] <http://localhost:7071/api/chats/{chatId}>
GetChatState: [GET] <http://localhost:7071/api/chats/{chatId}>
PostUserResponse: [POST] <http://localhost:7071/api/chats/{chatId}>
If the local debugging would work, we will move on to publish to Azure.
[Publish to Azure]
Use the following methods to publish your project:
Visual Studio Code: Use Azure Extension to publish.
Visual Studio: Use built-in publish profile. Also please note to enable SCM Basic Auth Publishing Credentials from Azure portal, as this is required by Visual Studio deployment:
Grant the user or function app managed identity Cognitive Services OpenAI User on Azure OpenAI resource. This is important as the platform will use it as authentication method to allow connection:
Result:
Create a new chatbot
Make conversations
Text Completion
Text completion allows Azure OpenAI service to extend or answer with given sentences. It’s commonly used with paper writing, story telling and many more scenarios. The below example will demo how to leverage completion APIs to perform text completion:
[Local Debugging]
Clone the GitHub repo, you may take a look at README.md to see detailed instructions.
Go to csharp-ooproc/local.settings.json. Replace values to match parameters in your Azure OpenAI endpoint. Also, remember to add setting CHAT_MODEL_DEPLOYMENT_NAME = <name of your deployed LLM>
In the TextCompletions.cs file, you can see 2 kinds of functions:
WhoIs: it will extract the value “name” based on the invocation URL format (whois/{name}), form the question (Who is {name}?) and send to Azure OpenAI service for text completion.
GenericCompletion: This function will directly take the prompt as input, and send to OpenAI completions API for text completion.
We will skip details of the test due to this has been covered in the README.md. If the local debugging would work, we will move on to publish to Azure.
[Publish to Azure]
Similar to chat section, please pay attention to the difference when using different IDEs.
Result:
WhoIs
GenericCompletion
Rag-cosmosDB
Azure cosmosDB product provides a good option to store previous chat history / company-level Knowledge Base. In this example, we will show how to leverage cosmosDB product to store the information, then use semantic search to locate and print required contents.
[Requirements]
Azure cosmosDB for MongoDB (vCore)
Azure OpenAI Service
Azure Function App
[How to work with Azure cosmosDB]
To work with Azure cosmosDB, firstly prepare the environment:
Clone the GitHub repo, you may take a look at README.md to see detailed instructions.
In the local.settings.json file, update the CosmosDBMongoConnectionString value to match the connection string from cosmosDB resource. Also, define the used embedding model in Azure OpenAI (appsetting: EMBEDDING_MODEL_DEPLOYMENT_NAME, default model: text-embedding-ada-002).
Please also check the README.md to fill the remaining parameter settings.
Then, we will follow the below steps to leverage cosmosDB:
Insert docs into cosmosDB. We will need a storage or equivalent service to host TXT/JSON file. This is the source where you can add or edit the content. Then invoke POST request like below:
My example: Mengyang Chen is a Support Engineer working for Azure App Service Team.
You can also validate the ingestion has been succeeded in the terminal logs:
Query by Prompt. By invoking POST request, we can receive the desired result via semantic search.
As you can see, it would give you the result and where to find this info. This will be helpful if users want to build custom Knowledge Base or store long-term memory.
Hope this blog will give a good start to leverage this extension, and if you want to ask anything related to this, please feel free to leave comments, and we would be glad to help.
Microsoft Tech Community – Latest Blogs –Read More
The send icon is not showing up on my emails on Outlook
Please tell me how to get my Send to show up on my emails
Please tell me how to get my Send to show up on my emails Read More
How to Apply Microsoft Edge Configuration Profiles to BYOD Users Without Device Enrollment?
I’m looking for some advice on managing Microsoft Edge settings for our BYOD employees. Specifically, I want to push configuration profiles to Microsoft Edge for users who access our environment through their personal devices, without enrolling these devices into Intune or our corporate environment.
The issue I’m facing is that these settings don’t seem to apply unless the devices are enrolled, which I’d like to avoid because the people will be hiring will not work with us long, at all.
These temp workers will be completely browser-based, nothing will be stored on the device and no programs will need to be installed. I’ve got my configuration profiles in place, as well as my conditional access. Conditional access and MCAS settings are applied, just not configuration profiles.
Sorry if this is a really dumb question, I’m really bad about missing the most obvious solutions. I really appreciate any advice and suggestions.
I’m looking for some advice on managing Microsoft Edge settings for our BYOD employees. Specifically, I want to push configuration profiles to Microsoft Edge for users who access our environment through their personal devices, without enrolling these devices into Intune or our corporate environment. The issue I’m facing is that these settings don’t seem to apply unless the devices are enrolled, which I’d like to avoid because the people will be hiring will not work with us long, at all. These temp workers will be completely browser-based, nothing will be stored on the device and no programs will need to be installed. I’ve got my configuration profiles in place, as well as my conditional access. Conditional access and MCAS settings are applied, just not configuration profiles. Sorry if this is a really dumb question, I’m really bad about missing the most obvious solutions. I really appreciate any advice and suggestions. Read More
shared OneDrive .xlsx file does not open on Chromebox
I used to be able to share a OneDrive .xlsx file via e-mail then open the file on Chromebox. Now when I click the file in the e-mail I get “Sorry, an error has occurred” message on the Microsoft OneDrive site on Chrome browser (on Chromebox). Has something changed in the last year where shared OneDrive files can no longer be accessed on Chromebox? I am signed into OneDrive on Chrome browser on Chromebox but still get same error message. Thanks.
I used to be able to share a OneDrive .xlsx file via e-mail then open the file on Chromebox. Now when I click the file in the e-mail I get “Sorry, an error has occurred” message on the Microsoft OneDrive site on Chrome browser (on Chromebox). Has something changed in the last year where shared OneDrive files can no longer be accessed on Chromebox? I am signed into OneDrive on Chrome browser on Chromebox but still get same error message. Thanks. Read More