From a Microsoft ML Studio and Visual Studio Code Fan

Photo by Mitchell Luo on Unsplash

I’m going to cut right to the quick of it — while I really like Azure ML Studio combined with (or without) Visual Studio Code for use with Python Jupyter Notebooks, if you just look at the steps to get up and running, Google Colab (Colaboratory) is hands-down the easier to use.

Assuming you have a Gmail (Colab) or Outlook (ML Studio) account, the steps to create a Jupyter Notebook in ML Studio are:

  1. Create an Azure account
  2. Create an Azure ML Studio resource in Azure
  3. Create a Compute Instance in Azure
  4. Start the Compute Instance
  5. Code away

Comparing that…


Opinion

The Python Programmer’s Perspective

Photo by King Lip on Unsplash

My journey in data science started during my Master’s in Data Science at Bellevue University, where I learned both R and Python. While I’m a pretty serious Python advocate, it’s not exactly the fastest programming language I’ve ever used (nor the slowest). But I’ve been hearing a lot about Julia’s speed, so I began to experiment with it. I even wrote a Medium.com article on getting started with Julia. I must admit, I left that article with no real impression of Julia because I hadn’t done anything useful with it.

But as I’ve said in previous articles, I like competition…


Insights from 11+ Years of Habitual Fitness

Photo by Victoire Joncheray on Unsplash

I grew up an athlete (competitive swimming) and at a very early age felt the bliss of exercise-induced endorphins. But as life progressed, exercise became less of a priority even as my younger self pulled on the strings of my subconsciousness to get back into it. Then working out started getting more time during the 2000’s before things really stuck in 2010, after which I tracked every single workout. That’s 5,444 workouts (01/01/2010 to 06/22/2021) and counting.

What does 5,444 workouts look like?

Here are some stats:

  • Workouts include running, cycling, swimming, walking, weight training, kayaking, landscaping (moving 5 tons…


When base spaCy needs a lift

Photo by Brett Jordan on Unsplash

Using spaCy for named entity recognition works well, but not in every situation, especially when it comes to person names. However, thanks to Hugging Face you can use Google BERT models as an ML engineer (not as a data scientist), to easily increase person NER accuracy.

DISCLAIMER: spaCy can incorporate techniques similar to what I’m about to describe, so don’t count spaCy out, this article is simply an alternative method of incorporate the technique.

Hugging Face describes itself as a community where we can “Build, train and deploy state of the art models powered by the reference open source in…


Use the Cloudmersive connector to make it easy

Photo by Olia Gozha on Unsplash

Reading a simple PDF file in Microsoft Power Automate can be made simple with the Cloudmersive connector; that connector has a free plan that you can use to determine if your automation will work, or if you’re just learning intelligent automation. I’m going to show you how to use Cloudmersive to convert a simple PDF from an email attachment to a text file in OneDrive.

As I began creating this tutorial, I already had my Power Automate account (you can sign up for a free trial at flow.microsoft.com). …


Learn the Ingest / Process / Produce design pattern to organize IA

Photo by Kelly Sikkema on Unsplash

Design patterns create structure, organization, and reusability, which can simplify any part of system development and I have such a pattern for intelligent automation called “ingest / process / produce.” This pattern not only describes a major type of intelligent automation but allows you to break such problems down more easily and consistently.

Once upon a time, back when the (design pattern) world was young, we just programmed systems (or maybe it was just me). I did not give explicit thought to the patterns that made up those systems even if at times I saw myself doing the same thing…


Avoid the nouns and get stuff done

Photo by Murilo Viviani on Unsplash

I’ve studied and worked with development methodologies for over 20 years and the best one I’ve ever used is Agile — the verb — which focuses on the key principle of The Agile Method, but without all those troublesome nouns.

In 1998, the team I was on hired IBM Global Consulting to help us with a project and they brought their development methodology called (in very typical IBM fashion) The Worldwide Software Design and Delivery Methodology (WSDDM). The base method had over 1,000 activities.

1,000 activities.

Then I progressed to the Rational Unified Process and then The Open Group Architecture…


How to (very) easily count instances of text within a string

Photo by Lewis Kang'ethe Ngugi on Unsplash

I do a great deal of work with strings — while most of what I do is via natural language processing (NLP), I frequently just have to count the number of instances of a substring that exist within a string and the Python count() function is an elegant, simple, and effective way of doing just that.

How it works is really the best part — you just use a string object and call count() giving it the string you want to count. By way of example, assume you have the following log file (adapted from a sample IBM log file):


Begin your journey with Community Edition

Photo by Rock'n Roll Monkey on Unsplash

Robotic Process Automation (RPA) can be a very good automation option to enhance human productivity often (but certainly not always) at a reasonable price point. There are several options available including UiPath but before you spend all kinds of money (UiPath is not inexpensive), you can see what it can do and if it can help your intelligent automation efforts by test driving the UiPath Community Edition.

To their credit, for educators and students (and anyone can be a student not just actual students), Community Edition is a free, fully functional version of UiPath Studio and here’s how you can…


Data Science saved me from myself

Photo by SOULSANA on Unsplash

I was tired of shaking hands and kissing babies and wanted to make more than money — I wanted to make a difference. I had spent over a decade in executive positions that I thought were what I wanted, but they left me an empty shell of a person and I just wanted to get back to doing something I loved doing. So, I reinvented myself.

That sounds really nice and theatric, doesn’t it? It’s mostly nonsense (the empty shell part is true). I really wanted to make more money by understanding data science as an executive, which is why…

Frank Neugebauer

Architect, adjunct professor, programmer, athlete, musician, intelligent automation leader. franklyai.medium.com

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store