Is anaconda necessary for Python

Python with anaconda

Python is the language of choice for projects in the field of machine learning and data science. Different versions of the language itself, but above all the thousands of packages and libraries required in this context, are difficult to control. Luckily it stands with Anaconda a tool is available that helps to keep track of things and to manage the project.

Anaconda describes itself as a data science platform, so it is a Python distribution with a focus on predictive analysis and scientific calculations.

Among other things, it includes:

  • the Conda Package Manager with the associated repository
  • the programming languages ​​Python and R
  • a cloud repository for sharing your own projects
  • a manager for virtual environments
  • Jupyter notebooks and other development environments such as Spyder or RStudio
  • the Anaconda Navigator

The Individual edition The current version Anaconda 3 is available for free and is aimed at solo developers. More extensive, but chargeable editions are available for teams or companies.

Installation of Python Anaconda

Anaconda is available for Windows, Mac OS, and Linux for Python 2.7 and Python 3.7, respectively. The ARM architecture is unfortunately not yet supported, so there is no official version for the Raspberry Pi. With the Miniconda (Link), however, a slimmed-down version is available. The graphical installer is 460 MB in size, the installation does not involve any major hurdles.

Anaconda Navigator

After installation, the Anaconda Navigator start, which provides an overview of all applications, packages and environments. On the left side there is a menu, the first point of which is “Home”. Additional applications such as the well-known Jupyter Notebook or the IDE Spyder can be installed here, or applications that have already been installed can be started.


Conda is a package manager similar to Pip, but goes beyond its scope of services. Pip is explicitly intended for Python packages, with Conda, on the other hand, you can also manage C libraries or R packages. The tool can be reached from the command line. For example, to install TensorFlow in the CPU variant, the following entries are required:

Conda will then download and install the required packages.

Virtual environments

In Python projects, it is preferred to work in isolated virtual environments. They allow the dependencies of different projects to be differentiated in the form of packages and modules. In principle, this is a directory structure that contains exactly the right Python version and the required packages for the respective project. The environment previously created and activated for TensorFlow called tf is now in the Anaconda Navigator under Environments to find. After selecting the environment name, the packages contained are listed with their respective versions on the right-hand side.

Python rocks!

python ipsum dolor sit amet consectetur, adipisicing elit. Iste atque ea quis molestias. Fugiat pariatur maxime quis culpa corporis vitae repudiandae aliquam voluptatem veniam, est atque cumque eum delectus sint!

Powered by Jekyll - Hosted by Github - Copyright © 2021 PythonBits