Welcome Message

Unus pro omnibus, omnes pro uno -- One for all, all for one

Welcome To Our Mutual Support Community Web Site
(Formerly Known As "The Intel Eliminati" - TIE)

Sunday, May 29, 2016

Anaconda Python For Fun and Profit

For anyone who intends to learn Python from scratch or enhance their programming skills by developing a small project, I highly recommend installing Anaconda Python provided by Continuum Analytics at this web site. The package is very complete and is provided free of charge.

When I was the systems software architect of the Electrical Validation department, we had many compatibility issues and endless number of package conflicts that stemmed from the absence of a reliable package management system. It was also very difficult to exchange code and track the dependencies of software components that were developed in python by more than 300 engineers, located in many sites around the world. During my investigation of the systemic problems that we faced daily and while looking for a way to solve these problems, I encountered Anaconda Python. After introducing it to some of my close colleagues, they became religious fanatics about using it. The nice thing about Anaconda, is that it is a completely self standing subsystem, which does not depend on any other system applications or system components, since it includes the "kitchen sink" for everything Python. In the Windows environment this has a significant advantage , because the Windows Registry is not used at all and consequently, limitations imposed by Registry mismanagement are non-existent.

The Anaconda package does not require any formal installation, you simply download the zip/tgz file and unpack in a folder of your choice. This applies to all OS versions, including Windows, Linux, and OS-X. Both 32-bit and 64-bit versions are supported. An internal package manager is used to maintain all software components current and compatible. The Anaconda environment also allows easy switching among different Python versions (2.x/3.x) if such an issue is of importance to your application.

The provided IDE is Spyder-- an excellent development and debugging tool that includes an iPython execution environment, which is very nice for "on the spot" test running and debugging of your code snippets. Anaconda also has very good support for developing Python "Notebooks" with Jupyter.

Since in my department we used a lot of scientific calculations, as well as "big data" processing, we found Anaconda to be the most friendly Python development environment that allows a developer to focus their efforts on development and production, rather than chasing compatibility problems.

A list of all the packages included in the most recent version of Anaconda is included at this link: Anaconda 4.0.0 package list .

Try and enjoy.

--Dr. Flywheel

No comments :