Data Science

SQL Server 2017 Machine Learning Services – Offline Installer Issue

by

Situation

You’re trying to install SQL Server 2017 Machine Learning Services onto an existing SQL Server 2017 installation.
You select the checkboxes for R en Python because that’s how you roll.
And off you go to the next screens!

Issue

That’s when you remember it… Your server isn’t connected to the internet!
Pretty normal, but in your enthusiasm you completely forgot that SQL Server needs to download some binaries for the R and Python components you so desperately want on your precious machine!

Luckily, the installer comes to your rescue and shows you where to download those binaries it needs.
Turns out however… This link only is for one R component and the installer won’t let you pass to the next screen!

Solution

Microsoft has a complete list of all possible components you could ever want to install while offline. From SQL Server 2016 RTM, over to SQL Server 2017 CTP 1 and up to SQL Server 2017 CU 1.

You can find the full list over at docs.microsoft.com: https://docs.microsoft.com/en-us/sql/advanced-analytics/r/installing-ml-components-without-internet-access

Hopefully, next time you’re installing ML Services, this will save you some time searching for why that “Next” button won’t become active.

 

SQL Saturday 642 – Sofia

by

I was lucky enough to get selected as one of the speakers for SQL Saturday 642 in Sofia this year.
Let’s do a quick review of my session and some sessions I visited.

Enabling Citizen Data Science with Microsoft

As a speaker you’re lucky enough to teach people what you know and experienced but also to get feedback from people in your session.
For the people who haven’t seen or heard my session before and who can’t make any sense from the abstract, I’ll slowly be blogging my entire session over the coming weeks. The very short version is: that while data science has become somewhat of a buzz word and a lot of people suddenly want that title. As BI Developers, Analysts, etc. it often is hard to know where to start. I guide you Microsoft’s 3 month long self-paced Data Science course, which covers theory and practice. And I cover the tools you need to get started.

I included the new Azure Machine Learning Services in my slides. Mainly because it’s new and actually very useful.
And one of the things I learned from my audience is that Microsoft’s announcement of these services is actually a bit confusing for people who are getting started. They now get the impression that this is something they NEED. While incredibly useful, when starting out it’s more important to get the basics right instead of trying to perfect the entire lifecycle.

Other very popular sessions were the “Database Continuous Delivery on the Microsoft Platform” by Gavin Campbell (blog | twitter) and “PowerBI for Rookies” by Miroslav Dimitrov.

Gavin Campbell talked about the theory, practice and different parts that make up a Continuous Deployment pipeline. From dacpac’s to version control to testing, building and onto automatically releasing your database.
Basically a must see session for everyone who’s developing databases.

Miroslav Dimitrov guided his huge audience through everything anyone would need to get started with Power BI.
From getting data, to creating a report and publishing a dashboard. Beyond that he talked about some security aspects and cool features like for example QuickInsights and publishing to the web.

Apart from these sessions there of course was a lot of food and enough drinks to be had by all the speakers who gathered on friday and saturday evening.  For me those tend to be the most memorable of an entire event because there’s always people at the table that I look up to.
This time I had the honor of sitting next to Dejan Sarka (blog | twitter) who’s advanced sessions at conferences and even pluralsight courses will teach something to even the smartest people (but also give them a headache because of the difficulty).

Lastly there’s the non-technical things I learn from people and speaking during dinner or the conference itself.

 

So thank you to the entire SQL Saturday Sofia team for organizing this great event and ensuring everyone had a great time.
For everyone who hasn’t attended one of these yet, start doing it! SQL Saturday’s, other conferences and user group meetings both virtual and real life are a good way to learn more and to get to know new people.

 

Microsoft Professional Program – Data Science

by

Last year, from oktober 2016 up to january 2017, I participated in Microsoft’s Professional Program, specifically the Data Science track.
It was only the 1st public iteration of the program but back then it already felt like a mature course.
This probably had to do with the fact that there had already been a private run of the course in the months before.
David Eldersveld (twitter | blog) was one of the participants in that original run and he gives you a high level overview on his blog.

In this post I’ll be going into a bit more detail and explain how I experienced the program.
In short, there was joy and there were tears.

Read on for the full story!

Read moreMicrosoft Professional Program – Data Science

I’m attending and presenting at… ALL THE EVENTS!

by

This year I’ve already presented at more events than I dreamt off at the start of the year (5).
And the invites keep coming in! (5 more!)
This is going to be a long read, a punishment from me to me because I slacked off in blogging in the past months.

So apart from being bored, there’s 2 reasons why you would want to read this post.
1) Interest in, but some fear of, attending and presenting at conferences, user group and community events in general.
2) You’re me from the future and you’re wondering about that amazement and the exciting feeling the young Jan had about these first events. You’re wondering about what fears you overcame and you want to look back at how it all started.

Expect to find the following:

  • UK Power BI Summit (2017/02/17)
  • Denver SQL Server User Group (2017/03/17)
  • Battle of the Beards (2017/03/29)
  • SQL Saturday Israel (2017/04/26)
  • Belgian Information Worked User Group (2017/05/09)
  • in short: 5 more events to find me at before the end of june!

So whichever reason you have, come on in and start reading!

Read moreI’m attending and presenting at… ALL THE EVENTS!

Presenting a webinar, not the same as a conference

by

Last saturday Back in January, I presented my first webinar for the Global Power BI usergroup.
It was a redelivery of the Personal BI to Personal Data Science session I’ve already given twice together with my colleague Kimberly Hermans (twitter).
Although there was no real negative feedback and even some positive feedback in private, I don’t think I did great.

I approached the webinar the same way as I do a regular presentation.
And boy oh boy, was I wrong to do it this way…

There’s all kinds of different and extra things to take into account compared to an in-person conference or usergroup presentation.

  • The software you’re using
    We used Google+ Hangouts and while I tried it out before the webinar, I didn’t prepare how I would be taking questions. That could’ve gone a lot better.
  • The microphone you use
    The microphone I used was the one that came with my phone. The quality wasn’t the best, especially combined with the room I was in.
  • The room you’re in
    Using a cheap microphone tends to be OK. But in the wrong room you’ll get a lot of echo or environment noise.
  • No or very limited interaction with your audience
    This one struck me the worst. I couldn’t interact with or read the audience which made me unusually nervous.

On top of that, because of conflicting schedules I had to present alone this time.
I thought I would be OK as I knew most of the data science stuff on a basic level.
But it also means that all the interaction and the dynamic that previously existed in the presentation was gone. No jokes, no natural tempo changes, no interaction between presenters.
In my opinion this was the main killer of the webinar.

What’s next?

UserGroup

The Virtual Global Power BI User Group is still organising monthly webinars. You can join or present yourself as well.
Or just participate in the usergroup via different channels like Twitter, Facebook, LinkedIn and our YouTube channel.

Personal

I’m embarking on a new webinar journey as well.
More news will follow soon.

 

Data science with Microsoft – Personal BI to Personal Data Science

by

How do you go from Personal BI to Personal Data Science?
Isn’t Data Science only for those rare unicorns that are smarter than most of us combined?
Can we even democratize Data Science within the enterprise?Personal BI & Data Science

Kimberly Hermans (twitter) and I presented on this topic last week, the 27th of November, at the Microsoft UK office. Jen Stirrup (blog | twitter) had organized a great event for the community there: “Data Culture Day London – Power BI Edition“.

This is a write-up of the presentation “Personal BI to Personal Data Science”, hopefully it will make you want to attend the presentation.
Currently you can still attend it at the next SQL Server Usergroup event in Belgium on the 8th of December.
After that date, take a look at my calendar to see if I’m delivering this session somewhere else.

Read on!

Read moreData science with Microsoft – Personal BI to Personal Data Science

Data science with Microsoft – Training materials

by

Today I’ll be guiding you through the, sometimes very busy, world of Microsoft training material.
We’ll put the focus on the training material for data science.

Expect everything you need to become fluent with Microsoft’s Data science solutions.

  • Free training courses with certificates
  • Free webinars & recordings of live sessions
  • Free (e-)books
  • Documentation & Learning Paths
  • Microsoft Virtual Academy

Read on for the good stuff!

Read moreData science with Microsoft – Training materials

Data science with Microsoft – An introduction

by

Let’s be honest, Microsoft isn’t a name that would traditionally be associated with data science.
But just as we’ve witnessed in other areas, they have quickly caught up!

in the last year we’ve seen the following appear on the Microsoft Data Science radar:

  • Azure Machine Learning
  • Power BI
  • Cortana Analytics Suite
  • Acquisition of Datazen & Revolution Analytics
  • Integration of R in SQL Server

Looking at it like this, it’s just a list like any other. Not even a big list.
The magic happens when we look at what this means for the developers, consultants and ultimately the business.

Azure Machine Learning

We now have the ability to create AND deploy predictive models in minutes using Azure Machine Learning.

AzureMachineLearning2 AzureMachineLearning

This opens up interesting possibilities where we can send data from SQL Server, a SQL Azure Database or just live from a mobile application or excel to gain insights

Power BI

If you still need an introduction to Power BI then you’re doing something wrong.
Contact me on twitter, LinkedIn or via the comments.
I’ll gladly talk you through Power BI and why you should be using it for almost everything you do with data in your company.
I mean that, get in touch with me. Now! 🙂

But on a more serious note, I’m going to be crude to Microsoft here.
A long time ago, Power BI started as an over-hyped and underwhelming experience. Everyone saw the potential this Excel stuff had but I’m guessing the experience most people had was similar to mine. That is, Power BI back then was a disappointment because of what we were expecting.
The one good thing it did have at one point was PowerPivot.

Skip forward to august 2015.
The Power BI dream had suddenly come true!
Most of the things we were expecting in the past suddenly were there, in a web service AND a desktop application.
AMAZING!

Skip forward 3 more months and Power BI has exceeded our wildest dreams.
I could literally fill books with all the great stuff the Power BI team has done and enabled for the community.
The Power BI API, a plugin for PowerPoint, custom visualizations, support on all devices, enterprise ready and a lot more all combined with a CRAZY pace of new releases!

If you haven’t used Power BI yet, skip all the praise and commercial talk, go download the desktop application and start working with it. Soon you’ll be an Power BI evangelist as well 😉

custom-visualizations-same-page-100622406-primary.idge[1]

Cortana Analytics Suite

If I had to summarize Cortana Analytics for anyone, I’d say it is basically Azure Machine Learning for predictive analytics combined with Power BI for a beautiful presentation of your data. And sprinkled on top are some of the most incredible and integrated services you can dream of.

Cortana Analytics is not really a product, it’s more a combination of several services that work really great together and form a solution to your questions.
It enables different scenario’s for any case you can think of.

Whether you have a scenario with real-time data analytics, (real-time) predictive analytics or you’re just in need of a data lake to fill with your data for analysis, Cortana Analytics is where you need to be.

This picture from Microsoft summarizes the Cortana Analytics Suite the best.
It shows you how different tools fit different purposes in the chain from data to insight to action.

Cortana-Analytics-Suite[1]

Acquisition of Datazen & Revolutions Analytics

Not much to say except: WOW!
I bet I’m not the only one who did not see both of these coming.

Datazen was already known for it’s mobile dashboarding solutions. It’s acquisition could only mean something big was coming for on-premises BI.
And it did, Microsoft announced at PASS Summit 2015 that Datazen would basically be integrated with SSRS to provide an outstanding mobile BI solution for those who must stay on-premises.

Revolution Analytics was widely known in the world of the R programming language.
Where R standard is limited to a single machine and the memory that machine has, Revolution Analytics provided a scalable solution.
How cool is that? So cool that Microsoft wanted it integrated in SQL Server 2016!
I’m sure that the R services in SQL Server 2016 are just a starting point. But imagine the possibilities from a data science perspective when you combine this with the columnstore and in-memory technologies.

splash[1] architecture[1]

Integration of R in SQL Server

This isn’t just R in SQL Server, it’s an implementation of Revolution R Enterprise in SQL Server!

No longer do you have to pull data to your developer machine, data can just stay in the database where it should be.
Combine this with columnstore indexes and the in-memory technology and you know that the data scientists are now drooling.
2015-05-14_22-39-58

Summary

Microsoft has improved so much in the last year, it’s as if it’s a whole new company.
Data professionals are getting a lot of shiny new toys and can expect a lot more solutions to be build end to end on a Microsoft platform.
Whether that platform is on-premises or in the cloud is up to the business to decide.

Data science is no longer unknown territory for people who work with SQL Server, it’s already on our doorstep.
On top of that, Microsoft’s Cortana Analytics solution offers incredible value and an ease of use I’ve never seen before with something like this.
It took me only an hour to set up a solution that parsed real-time sensor data, combined it with reference data in a database and then show it on a Power BI dashboard.

One thing is sure, you can expect some interesting blog posts in the feature.
Not only from myself but the entire SQL Server community!

When you’re ready, move on to this list of training materials I compiled for you. Let’s get started!

Cortana Analytics Suite – 2015/9 Workshop Videos

by

With all this news about Power BI it’s easy to forget about the important role it plays in the Cortana Analytics Suite.

Most of us couldn’t attend the Cortana Analytics Suite workshop in Seattle but that doesn’t mean you have to be sad!
Now you can watch all the videos from the comfort of your own chair!

Wait, you don’t know about the Cortana Analytics Suite or the workshop yet?

Quick, read on!

Read moreCortana Analytics Suite – 2015/9 Workshop Videos

Opendata and citizen datascience

by

Data analysis and visualizations are the most useful end products as BI professionals and even data scientists. They give actionable insights to the end user.

With all the data initiatives and people working with it, there are now a lot of examples of government open data being used to better the community.
But I’ve yet to see a lot of  Freedom of Information Act (FOIA) datasets being used or visualized.

This post uses an enriched dataset about the deaths in California Police Custody during the 2013 – 2015 period acquired using the FOIA.

Interested in getting the enriched dataset and analysing it yourself?
Read on!

Read moreOpendata and citizen datascience

%d bloggers like this: