Azure Skåne AI Day!

AI is becoming a more common topic for Azure Skåne User Group (#AzureSkane,) we have previously hosted several AI events, and will again Welcome you to yet another AI flavored event by Azure Skåne User Group!

We are proud to present a full packed day with Knowledge sharing from vastly experienced profiles.

See full schedule details here:

https://azure-skane-ai-day-2020.sessionize.com/

Azure Skåne AI Day will this time do a hybrid setup with livestream physically from FooCafe and feature speakers that are present physically and online. *Online version will be in big room with small number of seatings.

Attending online find the link here:

Attending physically please find the address here: Foo Café – Media Evolution City Stora Varvsgatan 6a · Malmö

*Attending Physically = Pizza and drinks 🙂

The day will feature sessions from variety of spectrums and also include hands on session to build something cool!

Agenda

• 10.00 – 10.15 Introduction and Welcome
• 10.15 – 11.15  Build intelligent search with Azure Cognitive Search and Azure Functions – Daniel Krzyczkowski
• 11.15 – 12.15  Monitoring service quality with Artificial Intelligence – Alexander Klein
• 12.15 – 13.00 Lunch
• 13.00 – 14.00 Azure Custom Vision in action – Goran Vuksic (Hands on!)
• 14.00 – 15.00 Transfer Learning for Deep Learning: From Custom Vision to TensorFlow & ML.NET – Luis Beltran

Speakers:

Alexander Klein: https://sessionize.com/alexander-klein

• Senior Business Intelligence consultant
• Twitter:  @SQL_Alex
• Blog:  consulting-bi.de (blog)

Goran Vuksic: https://sessionize.com/goran-vuksic

• Twitter:  @gvuksic
• Blog: https://www.linkedin.com/in/goranvuksic/detail/recent-activity/posts/ (blog)

Sessions:

Build intelligent search with Azure Cognitive Search and Azure Functions – Daniel Krzyczkowski

Nowadays many solutions provide search functionality which is so simple for the user but can be challenging to build especially when there are different document types and sources. This session will walk you through the intelligent search solution built to easily extract relevant content from the documents and data sources using Azure Cognitive Search and Azure Functions. We will discuss some concepts related with Azure Cognitive Search and enrichment pipeline skillsets. The demo will cover how to set up Azure Cognitive Search, how to enrich its search pipeline with custom skills using Azure Functions, and how to display results in the structured form.

Monitoring service quality with Artificial Intelligence – Alexander Klein

Since you often have the task of document recognition, text evaluation or image rating in Office 365, I show you the link to Cognitive Services, where you can generate various added values for your Office 365 project using image and text recognition methods. I use function calls for the direct call of the Cognitive Services and show you how you can directly access external API (of the Cognitive Services). The example in this session is about the evaluation of the service quality of the #SQLKellner. It explains step by step how to assemble all components to get a “Service AI Power App” that runs on every smartphone or tablet. The list of ingredients includes Cognitive Services (AI), Power Apps, an Azure SQL DB, a Logic App and Power BI.

Azure Custom Vision in action – Goran Vuksic (Hands on!)

In this hands-on workshop, we’ll make a deep dive into the Azure Custom Vision and we’ll train custom models for classification and object detection.

Through this workshop you will:

– learn Azure Custom Vision in details,

– prepare and tag images for model training,

– train models for classification and detection,

– retrain and improve you models,

– and more.

Please note: In order to follow this workshop, Azure account is required, if you don’t have one you can create it here for free: https://azure.microsoft.com/en-us/free/

Transfer Learning for Deep Learning: From Custom Vision to TensorFlow & ML.NET – Luis Beltran

Transfer learning is a machine learning technique in which a model that was developed for an initial task serves now as the starting point for a model on a second duty. It is quite useful in Deep Learning since compute and time resources are limited, so you a pre-trained model can be used as an input for a computer vision or natural language processing task.

Let’s demonstrate how Transfer Learning works in ML.NET by exploring the following scenario:

– Firstly, an Azure Custom Vision image classification model that uses the Open Images Dataset is trained, published and exported to TensorFlow.

– Then, transfer learning is applied to this model using the ML.NET Image Classification API in order to create a new, custom deep learning model to identify specific image categories. All the knowledge gained when solving the initial classification problem is useful for shortcutting another training process and solve a second classification.

*If time allows it, I can also include how to publish the model for external usage, either from a mobile application or website.