Hack Together showcase - Fabric Flashcards
Hi Folks, Alvaro here. Yes, I work with Paul, I won't be submitting my project to the Hack Together, but that doesn't mean I cannot share some inspiration with you! 😊 So I asked myself what could be a cool way to learn more about Fabric, maybe even get prepared for the DP-600 cert, while having a Hack Together type project? Enter Fabric Flashcards: Project Idea Create a set of study flashcards, based on the Microsoft Fabric documentation. The questions and answers should be generated Gen AI (Azure OpenAI). Each answer should link back to the Learn Module so I can keep learning more on the topic. So I grabbed the learn Markdown files from modules like here https://github.com/MicrosoftDocs/learn/blob/main/learn-pr/wwl/introduction-end-analytics-use-microsoft-fabric/includes/1-introduction.md and then crafted some prompts to start extracting questions from the different modules. Let me show you just a snippet of the code: def generate_questions(self, text): system_message = """ You are an assistant designed to help people learn from tutorials. You will receive a Markdown document, and extract from it pairs of questions and answers that will help the reader learn about the text. Questions and answers should be based on the input text. Extract at least 5 different pairs of questions and answers. Questions and answers should be short. Output should be valid JSON format. Here's an example of your output format: [{"Q": "What is the name of the assistant?", "A": "Learn Assistant"}] """ user_message = text return self.call_openai( self._deployment_name, system_message=system_message, user_message=user_message ) So as usual iterate over the Markdown files, and then use some prompt like that one to start populating a dataframe, with questions and answers. Once I got my data all formatted, then I created a React application, that you can use to play with the Flash cards on a mobile phone https://fabric-flashcards.com/ So you get the front of the card: And the back with the Answer One step further: Printed Flashcards So I thought it was cool to get the flashcards in printed format, but this meant solving some challenges, like how to link back to a learn module from a piece of paper? Well, QR codes exist for a while right? I wrote an application that would go through all the learn modules URLs and generate the respective QR code. Keep in mind when working with GenAI, you have to incorporate notions from Responsible AI (RAI). In this case means providing the user of the cards some way to check the source of the information they are learning from. Also, it becomes a cool UX detail. So that ended up looking pretty cool (thanks to a friend that helped me with the printed design!). What's next? My goal over the upcoming weeks is to create a nice walkthrough that can show you how to create the flashcards using Fabric and Azure OpenAI. In this way we can get the skills not only to work with Fabric, but also to integrate it with other aspects of Azure, where in real life projects, that's where your data might live. So we can learn from using Lakehouse both to store the original Markdown files, but also to store the generated QR codes. In your company this could be images from your different products. Also the idea is to learn to generate PDFs with the help of GenAI, but always based on your data. I guess for all this I end up using SparkML with notebooks in the data science experience of Fabric. We'll see. So what do you think? Please share your thoughts, and thanks for reading this far! Also, checkout this video I've made demonstrating the printed flashcards
Business central - How to change unit of measurement conversion rate under Item
Hello! I have my production BOM improved and vendors changed from time to time. All these changes will make the item unit of measurement conversion rate changed. How to change unit of measurement conversion rate under Item in Business central? Right now the system seems preventing me to change the conversation rate. Thank you!
Disappointed with Dataflow Gen2
I have found that Dataflow Gen2 which uses Power Query doesn't work the same way that Power Query in Power BI and Microsoft Excel. It times out when previewing or refreshing if the query doesn't complete in 10 minutes. Query previews that don't take quite as long in Power BI seem to timeout pretty easily there. Another issue I found is that Dataflow Gen2 that a column header title that has an added space character causes it to produce an error when publishing while it doesn't seem to mind it while previewing. Needless to say. I am extremely disappointed in it and will have to a find another method to load the data into the lakehouse I created for the event.
