Software needed for the course #

Throughout the course, it will be common to use some software for teaching and interactive purpouses (after all, it’s a computer science course). If you are not very used to tweaking your computer, then it might be hard sometimes to understand what to do. This page tries to make a comprehensive list of all the needed software for following easily the course.

Different OSs and Software Versions

This guide will differ in some points depending on the operating system that you might have. When that will be the case, you’ll see different tabs depending on the OS that you use. Rest assured that each software treated here will be compatible with your OS.

Since this guide can’t be updated every time a new version of a particular piece of software comes out (unless a major release with ground-breaking features gets published), we are trying to provide a version-agnostic guide that should work for every version. If you notice some mismatches between the screenshots, that is because this guide was written in more than one day, so some programs got updated in the meanwhile, but you don’t have to worry about that

Python (for Programming) #

Used for some of the courses in ACSAI, Python will be a very handful software, so it’s important to set it up properly. You’ll mainly need it in the following courses:

• Programming (first year, first semester);
• AI Lab (second year, second semester);
• Machine Learning (second year, second semester).

During the Programming course, you’ll be given some instructions on what packages you need to use and what to install. Here, we’ll break it down a bit and better explain what does what.

For the Programming course you’ll mainly need 2 things:

• a fresh installation of Python, whose version must be ≥ 3.7;
• a text editor (for coding, not like Word, LibreOffice or Google Docs) of your choice;

In this guide we won’t focus on the text editors, but only on how to install Python and the necessary stuff. The professors will suggest you to download Python through Anaconda, which offers in a single installation a GUI application for managing Python and its packages, Conda’s virtual environments (which will be useful later), the Spyder text editor and a lot of other tools.

Difference between Anaconda and Conda

In this guide, you’ll often find the terms of “Anaconda” and “Conda”, which represent two different things:

• “Anaconda” represents the full package provided by anaconda.org, which comprehends the Conda package manager and other programs;
• “Conda” is the package manager, which will be, in some cases throughout this guide, an alternative to the package managers of some programming languages (in most of the cases, it will replace Python’s PyPi package manager, also known as pip).

Now, before installing Conda, you have different roads ahead: since the whole package provided from Conda contains a lot of softwares, you may not want to install all of it (for example if you have an old computer or not a lot of space on your hardisk); perhaps you already have installed Python for other projects, and you don’t want to redo everything just for Conda. By itself, Conda is necessary for some packages, so here we’ll explain how to install Python in 2 different ways:

1. with the full version of Anaconda
   If you have enough space on your computer, and you don’t mind downloading some extra stuff, opt for this option;
   
   
2. with the minimal version of Anaconda (called Miniconda)
   If you don’t want to download the full package of Conda, and you feel a bit confident with the terminal, then this might be a good option for you.

1) Install through Anaconda #

1. Head over to the Anaconda download page and, depending on your operating system, download the install file. If you are on macOS, then you’ll have two download options depending on your CPU (if you are not sure, check the CPU model by going on About this Mac):
   • if you have a Macbook that came out after 2020 with an M* CPU (like M1, M2, etc…), then download the “64-Bit (Apple silicon) Graphical Installer”;
   • if you have a Macbook that came with an Intel CPU, then download the “64-Bit (Intel chip) Graphical Installer”;
2. Execute the installer (as an administrator on Windows and as super user through sudo on Linux) and follow the procedures. If everything went smoothly, Anaconda will be installed on your computer;

Windows: Add to PATH variable

On Windows, the installer will ask you, before the end of the installation, if you want to add Conda to the PATH environment variable.

What does it mean and what does it do? The PATH environment variable is a Windows variable which stores some locations on your disk where the applications are usually saved. This helps the OS to know where to find the binaries of each application. This is not only used on Windows, but also on UNIX systems (so macOS and Linux). Anaconda by default recommends you to not add it to the path.

This is done in order to have a separate console just for Conda, in order not to mix it with the standard terminal. However, this might create more confusion, as you would always need to open the Conda console each time that you’ll need it. By adding Conda to the PATH variable, Windows will be able to use Conda also from the Command Prompt and from the Powershell.

TLDR: if you want to be able to use Conda from anywhere in your PC, add it to the PATH variable.

3. Check that Conda is installed:

Windows

Open the menu, and scroll down on the installed applications until you find a folder called “Anaconda3 (XX-bit)” (where XX stands for either 32 or 64, depending if your operating system is a 32 or 64 bit one): open the folder and click on the Anaconda Prompt application. After it opens, execute the following command to test that everything is working fine:

conda --version

If everything goes well, you should see the following output, so the name conda and the currently installed version (in this case, conda 24.1.2)

TODO: insert screen of Anaconda prompt

Linux / macOS

Open the terminal, and type the following command to test that everything is working fine:

conda --version

You should see the following output, so the name conda and the currently installed version (in this case, conda 24.1.2);

4. And that’s it! If you managed to arrive this far, this means that you installed without problems Anaconda and Conda. You can proceed to the next section.

2) Install through Miniconda #

1. Head over to the Miniconda download page and download the installation file;

Windows

no

macOS

2. Depending on your CPU, choose your installer (either in the .pkg or .sh format):

   • if you have a Macbook that came out after 2020 with an M* CPU (like M1, M2, etc…), then download the “Miniconda3 macOS Apple M1 64-Bit pkg/bash”;
   • if you have a Macbook that came with an Intel CPU, then download the “Miniconda3 macOS Intel x86 64-Bit pkg/bash”;

   In this section, we’ll use the .pkg file, which is the graphical installer. If you prefer to use the .sh file, then follow the Linux instructions (since they are both UNIX based OSs, the terminal instructions are equivalent);

3. Once you downloaded the installer, run it and go through it:

1) First page of the installer

3) Wait until it finishes installing

<—>

2) Choose where to install Miniconda

4) You’re done! Go forth to the next step

4. Open the terminal, and type the following command to test that everything is working fine:

conda --version

You should see the following output, so the name conda and the currently installed version (in this case, conda 24.1.2);

Linux

2. Open a terminal, and cd to the Downloads folder (or where you downlaoded the install script). In our case, it will be under ~/Download:

cd Download

3. Execute the script with sh:

sh ./Miniconda3-latest-Linux-x86_64.sh

4. Once the script launches, press ENTER and scroll through the license (if you want to skip it, press Q) and accept it by typing yes;
5. If you want to install Miniconda under your home folder press ENTER, otherwise specify a new folder;
6. After pressing ENTER, Miniconda will unpack its resources in the folder specified above. After it, it will ask whether you want Conda to start each time you open the shell. If you want it to be always available once your shell starts, then type yes.
7. Open the terminal, and type the following command to test that everything is working fine:

conda --version

You should see the following output, so the name conda and the currently installed version (in this case, conda 24.1.2);

Installing the required packages #

While in the Programming course you’ll only need the standard library, for running and testing the homework you’ll need some extra packages. This is where Conda comes into play: it’s needed to download all the necessary stuff for running the tests on the homeworks. We’ll here provide a list of all the packages needed, alongside what they do and how to install them. The original list is available on the Q2A site.

All these packages can be installed in one go with the following command. This comman must be run either in the Conda Prompt (if you are on Windows and did not integrate Conda with your main shell) or in the shell/Terminal (on macOS or Linux):

conda install -c conda-forge \
    ddt pytest-timeout stopit pytest-profiling radon typeguard pandas

You can update all the packages installed with Conda in one go by executing the following command:

conda update --all

Regarding Spyder

While we said that we won’t cover a section over which coding editor is preferrable for the Programming course, we will go over how to install the Spyder IDE (Integrated Development Environment) because it will be the editor used by the professors during the lectures and it will also be the editor that you’ll have to use at the exam!

Spyder and all its dependencies can be installed with

conda install -c conda-forge spyder

In order to update it, you can use the same command for updating all the packages installed through Conda

Java #

Used in the Programming 2 course, Java is a fundamental part when it comes down to learning Object-Oriented Programming (OOP). Unlike for Python and all its dependencies, Java is pretty straightforward to install:

1. Head to the Java download page and download the installer (it will automatically detect the right installer for your OS);
2. Execute the installer and go through the installation process;
3. That’s it! You installed Java!

You can check that you installed Java without issues by opening a terminal and typing

java --version

You should see a similar output:

The professor will suggest you to also download IntelliJ IDEA, JetBrains’ Java IDE, which is one of the most advanced Java IDEs. There are two editions of IntelliJ IDEA: the community edition and the ultimate edition. The community edition is free for everyone for personal use only (academic use is fine), while the ultimate edition requires you to pay a license for it. Luckily, it’s possible to get a free license for all the JetBrains products while you are a student.

RARS #

Used in the Computer Architecture Unit 2 course, RARS is a RISC-V processor emulator, which allows you to write and run Assembly code for the RISC architecture. It’s possible to download the program at this link. Since it’s a Java application, you’ll need Java installed on your computer.

Warning

On RARS, whenever you want to save or open a file, the program will always open the file explorer on the location where RARS is saved. So for instance, if I saved RARS under ~/Applications/RARS.jar and my file is at ~/Documents/RISCV/my-file.asm, then when I open RARS and try to open my-file.asm, the explorer pop-up will be on ~/Applications.

In order to avoid this, the best strategy is to put RARS in a parent folder of the folder where you keep your Assembly files. So for instance, it could be put inside ~/Documents/RISCV

There are two available versions of RARS, a light themed one and a dark themed one. They are both working fine, and have the same capabilities. If you want to download the dark themed RARS, remember to download also the theme.properties file and place it in the same folder where you will place RARS (this is not required for the normal version of RARS). Here are the download links for both versions:

Download link for RARS	Download link for dark RARS

Python (for AI Lab) #

For the AI Lab: Computer Vision and NLP course our beloved Python will be needed again, only that this time we will need to install some new packages. Chances are that, after the Programming course, Conda and Python are still there on your computer. If that is not the case don’t worry, you don’t need to install them again. If they still are there on your computer, then you might want to take advantage of Conda’s virtual environment.

A virtual environment is a Conda feature that allows you to install Python’s packages and libraries in a separate and protected enviroment. This is so that you can have different configurations of packages that won’t clash between each other. Conda is one of the systems, alongside pyvenv, that allows users to make virtual environments.

If you don’t want to take advantage of Conda’s virtual environments, just go to the end of the section, where you’ll be able to copy a command for installing all of the needed packages for the course.

Creating a Conda environment #

1. In order to create a Conda environment, you’ll need to have installed either Anaconda or Miniconda. If you haven’t installed them, you can follow either the guide for installing Anaconda or the one for Miniconda;

2. Open either a Conda shell (on Windows) or a Terminal (or macOS or Linux) and type the following command:

   conda create --name <env_name> -y
   

   where <env_name> is a name that you will give to your environment. A good name could be ailab for instance, but it’s up to you to what suits you best;

3. Now, in order to use the environment, type the following command in the terminal:

   conda activate <env_name>
   

4. When a new environment gets created, it contains nothing. We must then install Python and pip, a package manager for Python’s libraries. We can install Python and pip with the following command:

   conda install python pip -y
   

Selecting the environment and default environment

The above command must be executed each time that you open up a terminal (if you run your scripts on the Terminal). By itself, Conda will start on the base environment, hence why you have to change it everytime. If you want to change it by default everytime, then you will need ot tweak up a bit with your terminal.

On code editors like Visual Studio Code, you can easily select via a menu which environment you want to use, and it will be remembered after closing the application. Even better, if you use Jupyter notebooks, they will ask you at the very beginning which kernel (or environment) you want to use, and also in that case your choice will be remembered by the editor.

**Select the environment on Visual Studio Code**

On single Python scripts, you can select your environment by doing the following:

1. Open a Python file, and go on the bottom right corner. There, you’ll find next to the Python button another button with a version of Python. Click that button;

2. A pop-up will open, asking you which environment you want to use. Select the environment that you prefer most.

---

On Jupyter notebooks, you can select your environment by doing the following:

1. Open a Jupyter notebook, and look for the top right corner. You’ll find a button that says “Detecting Kernels”: click on that button;

2. A pop-up will open, asking you which kind of kernel you want to load. Click on Python Environments;

3. Here, you’ll find a list of all your Conda environments. Select the environment that you want to use with the notebook;

4. And voilà, you are there! You can check that your kernel got loaded successfully by looking at the button of Step 1: if it now displays the name of your Conda environment, then you’re set!

Installing the required packages #

In order to install all the needed packages, you can use the pip package manager with the following command (if you haven’t installed pip, run on your terminal conda install -c conda-forge pip -y), which will install everything in one go:

pip install numpy matplotlib scikit-learn opencv-contrib-python seaborn nltk

Later on in the course you will also need to use Pytorch. In order to install it, then you must refer to Pytorch’s download page, since it will auto detect your system hardware and will thus indicate what’s the best setup for you.