Installation

Cloud-based platforms like Google COLAB or others are very useful for teaching and demonstration purposes, because the effort to set up a working environment is minimal.

Nevertheless, they always require an internet connection and in many cases a login step. Also, some of them are not entirely Open Source.

Therefore, in this section we briefly describe one of the possible approaches that can be used to install some tools useful for developing with Python.

Actually, there are many possibilities to install it, and many systems come with Python already installed. Installing all the required libraries can be time consuming. To speed up the installation, there are some “bundle” distributions that are “self contained” and contain the main scientific/graphical packages. Hereinafter we provide some details about Anaconda and Miniconda.

Note

We suggest to install the Miniconda package. However, if you prefer, please feel free to opt for other installation options. For example, the Mamba package manager seems to be a valid and faster alternative to Conda and Miniconda. To install it, you can follow the link for the Mamba installation

Miniconda

Miniconda is a subset of Anaconda project that contains only a reduced number of packages pre-installed. To install it, download the version corresponding to your system at https://docs.conda.io/en/latest/miniconda.html. Hereinafter we describe the installation procedure for MS Windows, but the procedure should be similar also for other OSs.

Download the right file and then execute it. Follow the installation prompt with the default options, that should be OK in most of the cases.

After the installation, you should be able to find, among the applications, something like “Anaconda prompt” or similar - open it.

There are some preliminary steps that are useful to set-up your Python environment. These should be valid for all the OSs. It is suggested to:

Check if on the right, at the beginning of the text line on your prompt, there is the string (base). In that case, you should type:
```
conda deactivate
```
If when you open the “Anaconda prompt” there is no (base) text on the left side of your prompt, this step is not required. Nevertheless, running it should not create any problem to your set-up.
Miniconda creates by default a virtual environment called “base”. Here I suggest to create a new virtual environment related to the course. For example, you can name it like the acronym of this course (here the acronym pygreen is used, but of course feel free to use any other name). Write on the prompt:
```
conda create --name pygreen
```
In this case we created the “virtual environment” pygreen.
Activate the new environment:
```
conda activate pygreen
```
Now you should see on the left, at the beginning of your prompt line, the string (pygreen).
Install the required packages:
```
conda install numpy scipy matplotlib spyder
```
(Reply y or simply hit <Enter> to the installation requests. In general, the option between square brackets [ ] is the one selected by default if you only hit <Enter>. Be patient.) These are the main packages that will be used during the course, but some other will be required and installed when needed.
Now you are ready to launch spyder, one of the available Integrated development environment (IDE) that you can use to develop with Python. From the prompt, simply write spyder and hit <Enter>. The latest versions of spyder should already be run in “background”. Otherwise, if spyder does not run automatically in “background”, you can use start /b spyder in MS Windows or spyder & in macOS or Linux.

Anaconda

Warning

Use this option only if you have a lot of free space on your computer (it requires at least 1 Gigabyte of space) and your computer is relatively recent. If you need a more minimalist installation, please have a look at the Miniconda option.

One of the most common pre-packed container of Python utilities and libraries is Anaconda.

To install Anaconda, do the following: download from this link the distribution suitable for your OS. I suggest you to download version 3.X (at the time of this writing, the default version is Python 3.11). Once downloaded, follow the instructions step-by-step. If you do not have any other Python version installed on your system, then you can keep the default settings.

To see if the installation was OK:

On Linux:: Open the shell, then write spyder (or in some cases spyder3) and hit <enter>.
On Windows:: From the start menu you should see all the installed packages. Launch Spyder.
On Mac OS X:: You can do like in Linux. Alternatively, double-click on the file Launcher.app stored in your home directory into the sub-directory anaconda, and Launch spyder-app.

In all the cases, after some time the working environment spyder should appear.

Linux “system” packages

Note

Use this option only if you are already quite familiar with Python and Linux. Otherwise, please consider installing using the Miniconda approach.

On Ubuntu you can use the package manager to install spyder, python3-numpy, python3-scipy, python3-matplotlib and python3-pandas. The procedure should be similar for other distributions. Please also note that the use of a virtual environment is strongly suggested.