3.1. Installation of Python#
Python is a powerful, versatile and extensible open-source programming language. It is widely used for artificial intelligence applications, interactive web sites, and scientific processing. As of May 2024, the Python Package Index (PyPI), the official repository for third-party Python software, contained over 530,000 packages that extend its functionality [1]. ModelFlow is one of these packages.
Python comes in many flavors and ModelFlow will work with most of them. Nevertheless, users are strongly advised to use either the Anaconda distribution of Python or the closely related MiniConda distribution.
Anaconda is a full-featured distribution of python that comes with a comprehensive set of pre-installed packages and tools for scientific computing and data analysis. It is best suited for users who want a ready-to-use platform with a wide range of packages and don’t mind the larger installation size.
MiniConda is a more streamlined distribution, providing only the essential components for running python. It offers faster installation and a smaller footprint, making it suitable for users who prefer a more streamlined environment and want more control over package selection. Features that are in Anaconda, but not installed by miniconda by default, can be added manually.
3.1.1. Which is better for me#
In general, Anaconda is preferred by users who value convenience and desire a comprehensive package ecosystem. MiniConda is preferred by users who seek a minimalistic setup and have specific package requirements. If you are already familiar with python, plan to use it with ModelFlow and have space limitations MiniConda is probably the best solution for you. For users new to python Anaconda may be a better solution.
The processes for installing Anaconda and MiniConda are very similar. The main difference is which installer is downloaded and then where the distribution is stored in the file system.
It is possible to install both distributions on the same machine without them interfering with each other.
Both are available for the Windows, MacOS and Linux operating systems and ModelFlow should work equally well under all three operating systems. However, only the Windows version has been thoroughly tested and was used in producing this manual.
3.1.2. Anaconda/MiniConda installation instructions#
3.1.2.1. Windows#
The definitive source for installing Anaconda under windows can be found here.
The definitive source for installing MiniConda under windows can be found here.
The version of ModelFlow and the examples in this manual were developed and tested using the 3.10 version of python.
Warning
It is strongly advised that Anaconda/Miniconda be installed for a single user (Just Me). This is much easier to maintain over time, especially in a professional environment where users may not have administrator rights on their computers. Installing “For all users on this computer” (the other option offered by the installers) will substantially increase the complexity of maintaining python on your computer.
3.1.2.2. MacOS#
The definitive source for installing Anaconda under macOS can be found here: https://docs.anaconda.com/anaconda/install/mac-os/.
The definitive source for installing MiniConda under macOS can be found here: https://docs.anaconda.com/free/miniconda/miniconda-other-installer-links/.
3.1.2.3. Linux#
The definitive source for installing Anaconda under Linux can be found here: https://docs.anaconda.com/anaconda/install/linux/.
The definitive source for installing MiniConda under Linux can be found here: https://docs.conda.io/projects/conda/en/latest/user-guide/install/linux.html.