Name: SAP Data Intelligence ML Scenario Manager (01) - Overview & Documentation
Uploaded: 2023-9-4
Duration: 401 s

Share feedback

Planned maintenance of Practice Systems

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>The SAP Data Intelligence ML Scenario Manager helps you to organize your data science artifacts and to manage your tasks in a central location. As a multi-faceted data science application, it is built around the key concept of the machine learning (ML) scenario, which may contain datasets, pipelines, and Jupyter notebooks. In this nine part video series regarding the SAP Data Intelligence Machine Learning Scenario Manager, Tahir Hussain Babar (Bob) gives an overview of the series, with 30 second highlights of each of the 9 videos. Documentation and Resources are also looked at the end of the video. 0:00 - Introduction 0:30 - Scenario 2:40 - Data set 3:00 - ML Scenario Manager 3:30 - Notebook (data scientist) 4:30 - Modeler (data engineer) 5:40 - Documentation <a href="https://sap.to/6057KrmJF" rel="nofollow noopener noreferrer" target="_blank">https://sap.to/6057KrmJF</a> Please note: 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. Also, this video includes automatically generated captions and therefore might misrepresent the spoken content or display incorrect product names</div></div>

SAP Data Intelligence ML Scenario Manager  (01) - Overview & Documentation

Welcome to the SAP HANA Academy. My name's Bob, and in this video series, we'll be looking at SAP Data Intelligence. In this video series, we'll be focusing on using the Machine Learning Scenario Manager. And what we'll cover is how to create predictions within the Machine Learning Scenario Manager and then expose them so that they can be accessed via an API. In this first video, we'll be going through an overview and we'll be looking at the highlights of the complete video series. In this playlist, you'll be experiencing the different steps of a typical data science project within SAP Data Intelligence. We'll begin with a data scientist persona. We'll connect to and then we'll explore a data source before applying machine learning algorithms to train a predictive model. In our scenario, the data scientists will then collaborate with IT to bring the machine learning model into an ongoing governed process to provide value to business processes. So you should imagine a scenario where a company specializes in organizing marathon races. Now, without a comprehensive predictive analysis based on historical data, an accurate automated prediction of marathon performance time is still a major challenge. So in our scenario as a data scientist we've been tasked with developing a machine learning model using SAP Data Intelligence that predicts the marathon performance time of all your runners. We'll do this by analyzing historical information such as half marathon times to predict the marathon's performance time for each runner. We'll then build a trained model which will then be exposed for inference or in order to make predictions which can then be used by business applications or mobile devices. Now the end result of our video series is that we're going to use mailman as follows. So I'm going to send the value in the terms of minutes which it takes a runner to run a half marathon and what will be returned will be the prediction of what the model thinks it will take to run a full marathon. In this example, if we have a runner who runs a half marathon in 135 minutes, we're predicting that it will take 297 minutes to run a full marathon. And as we're accessing this API using Mailman, any business application or mobile device can also access this type of data, including SAP Business Technology Platform. So in the next video, what we'll do is we'll start off by looking at our data set. Now, we've only got 117 rows of this data, but we'll be able to make predictions for a wide variety of half marathon minutes. We'll be using the Metadata Explorer to both investigate and profile this data. When we've looked at the data, we're going to really focus in this video series on using the Machine Learning Scenario Manager. And what we're going to do next is go through an overview of the front end and look at the basics of creating notebooks. And by the end of the video series, you'll hopefully understand the concepts of notebooks, pipelines, executions, models and deployments. So after the overview of the Scenario Manager, what we'll do is we'll start building notebooks. And initially what we're going to do is connect to our source data and just read the source data from our semantic data lake. We're going to read that CSV file within our Jupyter notebook. This ensures that connectivity works. We'll also look at performing some basic data analysis of our source data. Now after we've done that, we'll then use the source data to train a linear regression model in Python with just a single predictor column. So based on how many minutes it takes a person to run a half marathon, the model will estimate how long it takes that same person to complete a full marathon. Now that we've got this model, what we're able to perform are some predictions. Therefore, solely in our Jupiter notebook, we're able to supply how long it takes someone to run a half marathon in minutes, in this example 140, and it will return the prediction of how long it thinks it will take that same runner to run the full marathon. Which in this case is 309 minutes. So moving from the data scientist persona to the IT's persona, everything now is in place to start deploying the machine learning model within pipelines for use in production. So we're going to look at how two pipelines are built. One of the pipelines is used to train the model and save it into the machine learning scenario. And here what we're going to do is highlight the key parts or the key operators within this pipeline and show how they configured. In showing you this, as well as delving into the Python, we're able to reuse a lot of the Python that we experimented with within our Jupyter Notebook. Afterwards, we'll then look at executing that pipeline. And this pipeline in this first example is going to train the model and also output some quality indicator metrics which can be used for further analysis. Now that we've built our model, we'll want to use it and to continually access it using the REST API. So we'll also add another pipeline to surface the model as a REST API in order to retrieve those predictions. And again, we'll be using Mailman. As you can imagine, in terms of documentation, we'll be utilizing help.sap.com. And if I go to the Help portal, all we'll have to search for is Data Intelligence and then select Data Intelligence Cloud. And you'll find a whole list of resources here in terms of getting started, machine learning, and also modeling. So, if you like what you see, you should move onwards to the next video where we look at the data source using the Metadata Explorer, and we have a quick look at connection management. The whole video series is about 10 videos long, each video is around 7 minutes, so we hope you enjoy it and thanks for watching.