The last time that I created a playground for experimenting with machine learning using Apache Spark and an InterSystems data platform, see Machine Learning with Spark and Caché, I installed and configured everything directly on my laptop: Caché, Python, Apache Spark, Java, some Hadoop libraries, to name a few. It required some effort, but eventually it worked.

Hi guys!

Portrait of Madame X, Gustave Caillebotte.

One of the features I like in InterSystems ObjectScript is how you can process array transformations in a specific method or a function.

Usually when we say "process an array" we assume a very straightforward algorithm which loops through an array and does something with its entries upon a certain rule.

The trick is how you transfer an array to work with into a function.

One of the nice approaches on how to pass the information about an array is using $Name and Indirection operator.

Below you can find a very simple example which illustrates the thing.

After many sleepless nights it's a pleasure to announce the newer, better, moderner ObjectScript compiler which implemented pretty much everything you ever wanted to have in modern ObjectScript:

• Design objective of this new compiler is to parse reasonable subset of current ObjectScript syntax which will look readable for stranger, and not scare them with 1 letter syntax. The good start for compiler was the old-good COS Guidelines from here https://github.com/intersystems-ru/cos-guidelines
• For reasons we mentioned above we do not parse 1 letter syntax. It's declared evil;
• We do not handle dotted syntax for the same reason - modern syntax with {} is proper replacement for dotted syntax blocks;

But we not only parse the modern ObjectScript syntax, we have implemented finally the long-standing request which we always dreamed about. Closures!

How Tax Service, OpenStreetMap, and InterSystems IRIS
could help developers get clean addresses

Pieter Brueghel the Younger, Paying the Tax (The Tax Collector), 1640

In my previous article, we just skimmed the surface of objects. Let's continue our reconnaissance. Today's topic is a tough one. It's not quite BIG DATA, but it's still the data not easy to work with: we're talking about fairly large amounts of data. It won't all fit into RAM at once, and some of it won't even fit on the drive (not due to lack of space, but because there's a lot of junk). The name of our subject is FIAS DB: the Federal Information Address System database - the databases of addresses in Russia. The archive is 5.5 GB. And it's a compressed XML file. After extraction, it will be a full 53 GB (set aside 110 GB for extraction). And when you start to parse and convert it, that 110 GB won't be enough. There won't be enough RAM either.

This is a continuation of my story about the development of my project isc-tar started in the first part.

Just having tests is not enough, it does not mean that you will run tests after all changes. Running tests should be automated, and when you cover all your functionality with tests, everything should work well after any change in any place. And Continuous Integration (CI) helps to keep the code and deployment procedure with as fewer bugs as possible and automates the routine procedures, like publishing releases.

I use GitHub to store the source code. And some time ago GitHub started to work on its own CI/CD platform and named it GitHub Actions. It is not widely available, yet. You have to be signed as a beta tester for this feature, as I did. GitHub Actions uses quite a different way how to deal with a build workflow. What is important that Github Actions allows to use Docker, and it’s quite easy to customize available actions. And interesting that GitHub Actions is really much bigger than any classic CI like we have in Travis, Circle or Gitlab CI and so on. You can find more in the official documentation.

I am just recently announced my project isc-tar. But sometimes it is not less interesting what’s behind the scene: how it was built, how it works and what happens around the project. Here is the story:

• How to develop this project
• How to test it
• How to release new versions for publishing
• And finally how to automate all above
• Continuous integration

So, I would like to tell all about it.

Keywords: Anaconda, Jupyter Notebook, Tensorflow GPU, Deep Learning, Python 3 and HealthShare

1. Purpose and Objectives

This "Part I" is a quick record on how to set up a "simple" but popular deep learning demo environment step-by-step with a Python 3 binding to a HealthShare 2017.2.1 instance . I used a Win10 laptop at hand, but the approach works the same on MacOS and Linux.

This code snippet is an example of a CSP page that has a dynamically reloading table:

ˮ This is one of my articles which was never published in English. Let's fix it!

1. Blockchain

As I am writing this article, Bitcoin costs less than one-fifth of what it used to be at the pinnacle of its success. So when I start telling someone about my blockchain experience, the first thing I hear is undisguised skepticism: "who needs this blockchain stuff now anyway?"

That's right, the blockchain hype has waned. However, the technologies it is based on are here to stay and will continue being used in particular areas.The Internet in general offer tons of materials describing the general usage of these technologies

It will demonstrate the wide range that is openend by making use
of the power embedded in Node.js and its adapter to Caché, Ensemble, Health,..*
Node / JavaScript have wide reputation to work as a WebSocket client.
By using the Caché adapter it becomes easy to control it and to consume the results as a
Client for WebSocket Servers and to collect the replies in Caché, Ensemble, ..

I used node-v6.16.0-x64.msi and cache610.node as cache.node

You provide a Global for input:

There are often questions surrounding the ideal Apache HTTPD Web Server configuration for HealthShare. The contents of this article will outline the initial recommended web server configuration for any HealthShare product.

As a starting point, Apache HTTPD version 2.4.x (64-bit) is recommended. Earlier versions such as 2.2.x are available, however version 2.2 is not recommended for performance and scalability of HealthShare.

Recently I needed a classmethod that returns annotation value based on a name of a activity.

As doing it at runtime seemed inefficient, I wrote compile-time utility that iterates over all business process activities and generates relevant code.

This code could be used in a variety of situations when you need to iterate over business process activities, just add it as a secondary superclass to your BPL processes.

Running predictive models natively in an InterSystems IRIS Business Process has of course always been the goal of our PMML support, but somehow never made it into the kit because there were a few dependencies and choices that needed addressing and answering. Anyhow, thanks to some pushing and code kindly provided by @Amir Samary (Thanks again Amir!), we finally got it wrapped in a GitHub repo for your enjoyment, review and suggestions.