Name: SAP Artificial Intelligence Onboarding : Overview
Uploaded: 2023-9-7
Duration: 714 s

Share feedback

<p><u><strong>Planned maintenance of Practice Systems</strong></u></p>

Planned maintenance of Practice Systems

Why create an SAP account?

Enhance your learning experience

SAP Learning

Choose your subscription

Manage your subscription

Live Sessions

Hands-on Practice

Certification

My Learning

Ready to start learning?

<div><div>In this video tutorial, Tahir Hussain Babar (Bob) shows how to onboard SAP Artificial Intelligence & how to setup pre-requisites. This 1st video gives an Introduction to SAP AI and an Overview of the architecture. Documentation is also shown. <br><br>00:00 - Intro<br>00:22 - SAP Artificial Intelligence Overview<br>02:42 - Architectural Overview<br>10:25 - Documentation<br><br>SAP AI in BTP ; <a href="https://www.sap.com/products/business-technology-platform/artificial-intelligence.html" rel="nofollow noopener noreferrer" target="_blank">https://www.sap.com/products/business-technology-platform/artificial-intelligence.html</a><br><br><br>Please note:<br>                     This video was originally published on the SAP HANA Academy YouTube Channel                      and therefore may contain references to SAP HANA Academy and YouTube.                      The tech world is always evolving and the SAP HANA Academy YouTube channel                      is set to close on December 31, 2023.                      At that time, references will no longer be valid                      and will gradually be removed when the videos are updated.<br>Also, this video includes automatically generated captions                      and therefore might misrepresent the spoken content or display incorrect product names.</div></div>

SAP Artificial Intelligence Onboarding : Overview

Welcome to the SAP HANA Academy. My name is Bob, and in this video series we'll be looking at SAP Artificial Intelligence. In this video series, we'll be looking at onboarding. And in this first video of the series, we'll be going through an overview of SAP Artificial Intelligence. If you go on SAP's own product page for Artificial Intelligence for SAP BTP, there's a good one-pager on how it works. And as you can imagine, because it sits on the SAP Business Technology Platform, you can utilize all the other services available to you within SAP BTP. Now in this technical series, we'll be looking at the specific AI solution and capabilities which are available from SAP. And I'll firstly cover the ones that we're not going to cover in this video series. So the first one we're not going to cover is SAP AI Business Services. And as it says, you can add intelligence to your applications using AI models pre-trained on business relevant data. Another option is SAP Conversational AI. And this is where you can develop and integrate chatbots with your SAP applications in a low code environment. We're not going to cover this. The main ones we're going to cover are SAP AI Core and SAP AI Launchpad. So SAP AI Core is a service where you can deploy and run all your AI models at scale, without compromising data privacy and cost effectiveness. Essentially there are two parts to it. You can train your models and then serve those models to multiple endpoints. So how can we manage these AI Core services? Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. Hello, welcome to my lecture. There's links to other tools such as SAP Analytics Cloud and SAP Process Automation, where you can analyze your AI data. You can also see some examples of what customers are doing. And at the very bottom, if you scroll right down, there's a list of resources which you can use to learn about the various AI solutions and services from SAP. There's videos, there's tutorials, and of course there's a whole list of articles and blogs as well. In terms of the architecture, and as we've already mentioned, we'll be building this on SAP Business Technology Platform. In the first five or six videos we'll go through the steps of onboarding SAP BTP, as well as setting up the various sub-accounts to have the correct rights in order to access the various services you'll need. Of course we'll be setting this up in a sub account and we'll be using Cloud Foundry. We're setting up the Cloud Foundry environment because in a future playlist we'll be developing some apps. But you don't really need it if you're just using SAP Artificial Intelligence. In our scenario in this video series we're going to be using three core components. Of course we've already mentioned the SAP AI Launchpad and this is a tool that you subscribe to where you can manage a number of your AI runtimes or SAP core services. It will allow various users to access and manage their AI scenarios. We'll also be using something called SAP AI Core, which of course provides an engine for running any AI workflows and model serving workloads. Now what you'll do is you'll connect your AI launchpad to your AI core services and you'll use something called an operations manager to carry out various tasks such as link to data sets, create scenarios, create executions and create models. So a third key component we're going to look at is the AI API which we're going to be accessing. We're going to be accessing it through Python using Jupyter Notebooks. So an example of using the AI API is that we will register a GitHub repository which is an operation within the AI launchpad and we'll be using an AI API via Python and via a Jupyter notebook to do this. Other examples of operations are using the AI API to create scenarios, configurations and also links to data sets. The way operations can be created using the AI API is via an API connector which will enable operations to be created with commands using that API. And even though we'll be using a Jupyter Notebook to create these commands in Python, you should be aware that there are other run times as well, such as curl. And of course, you can access these commands or operations using Mailman. But again, in this video series, we're going to be concentrating on using Jupyter Notebooks to execute Python. We won't have an example of curl, but at the end of the video series, we'll have an example of using mailman to make predictions. When it comes to configuring how you want to train your model or serve your model, you'll be utilizing something called workflow templates and serving templates. So essentially, the first step is to always train your model based on the available data set. And then you'll want that model to be served for further analysis. So that's why we have the concept of workflow templates for training, and serving templates for inferring the data from those models. These workflow templates, whether you're modeling or serving the data, can be stored in a GitHub repository as a YAML file. So this concept of continuous integration and continuous delivery enables the ability whenever code changes, all the various systems will be aware of those changes. Welcome to my lecture. Today we're going to be going over how to install a Kubernetes cluster. This synchronization happens every few minutes. So as part of the on-boarding video series, I'll be showing you how to set up your GitHub repository. And with SAP Core AI administration, what we'll do is we'll create a link to your GitHub repository. I will create an application which essentially is a folder within your GitHub repository, which will contain these workflow template files. Now these workflow templates are used to train your model. But there's also a concept called a serving template which is actually can be stored in the same folder within your GitHub repository and these are used to serve your models. All this really means is that when you want to make predictions based on your trained model, you'll access a kind of a web server or model server via a URL in order to make those predictions. In terms of processing your data, we'll be using Docker, and Docker will be used to store Python code for training and serving in the form of something called a container, which is a portable environment. These Docker containers represent the environment where your code will be executed, and a Docker container is the place where your artifacts, whether they're data or models, will be copied from or to the configured object store. So because of this, the Docker repository will need to be connected to your SAP AI Core service for developing, shipping and running applications in these containers. Of course, what it ultimately means is that we'll be using something called Kubernetes, which is used by SAP AI Core to orchestrate the container platform. Kubernetes is great because it assures scalability, monitoring and cross platform deployments. So it's a great utility to use alongside Docker. Again, within the onboarding videos, I'm going to show you how to connect your Docker repositories to your SAP AI Core service. When it comes to data storage, there are some different options. And the type of things that these object stores host include things like your input data sets and your trained models. In this video series, we'll be using an S3 bucket. So I'm going to show you how to connect to that S3 bucket to access your input data set. The idea of using these hyperscale storage mechanisms is that you have the ability to massively expand the size of data that you can store under a controlled and efficiently managed software defined framework. Finally, once you've built, trained and served your models, you'll want to look at building applications that enable you to supply data to your model in order to make predictions. Now these apps can be built either within the SAP Business Technology platform, and of course because you're accessing these AI APIs they can be built and hosted elsewhere as well. As well as building apps to submit data to your AI core models for inferencing using API endpoints, you can also use apps to extend S4 HANA, and an example of that would be to trigger a maintenance order in S4. Additionally, you can use planning and visualization tools such as the SAP Analytics Cloud to explore and analyze the data from these AI models. In terms of resources, well, we'll be using help.sap.com. And of course, rather than show you the code, I'll be performing the steps myself. But I just want to point you to two main documents. If you do a search for AI Core within the SAP Help Portal, of course we'll be using the SAP AI Core service, and there's lots of documentation available on this service. So you've got basic information at the top, but then you've got information as well as getting started on development and also security. What you'll notice on the top right hand side as well is that you've got this related application, which of course is the SAP AI Launchpad. And we've got similar document sets there as well. So if you like what you see, you should follow the next series of videos, which again go through the process of onboarding yourself within the Business Technology platform, enabling the various AI servers, and then setting up all the prerequisites needed to do artificial intelligence using SAP. Again we'll be creating GitHub repositories and Docker repositories as well to do this. We hope you enjoy the video series and of course thanks for watching.