These are some very general instructions that may or may not work for your own particular machine. If you're having problems feel free to ask and someone will try to give you a hand.
For python on debian derived distros:
sudo apt-get install python
And to get pip:
sudo apt-get install python-pip
This may fail to get the correct Python version on some older distro releases. In this case using a python environment manager like anaconda or pyenv is suggested.
You can download an installation script for either anaconda or the lighter weight miniconda from (here)[https://www.anaconda.com/download/]. When prompted at the end of your installation, let anaconda modify your .bashrc file.
Re-load your bashrc file using
source ~/.bashrc
And you should be up and running.
Pyenv manages a separate Python environment on a per-subdirectory basis. This makes development and maintaining different versions of Python across different directories much easier than swapping between environments using conda.
For ubuntu based distros:
sudo apt-get install git
Other distros should have a similar package in their package manager.
We're going to need a few Python packages for today, they are:
ipython
numpy
matplotlib
seaborn
jupyter
qiskit
pip is the default Python package manager. We'll be installing to the local user's Python environment in order to avoid any permissions issues.
pip install --user ipython matplotlib numpy jupyter
This is generally a good idea to avoid conflicts with any other package managers you might have.
If you've used an anaconda or miniconda install you can also use the conda package manager. This shouldn't conflict with pip but sometimes has more recent releases (which may include essential bugfixes).
conda install ipython matplotlib numpy jupyter
Homebrew is a package manager for Mac. Using this will likely make the installation process much easier, but you can still do it manually if you want.
Obviously Python is important for a Python workshop, you can check if you already have Python by opening a terminal using 'spotlight' (Command + Spacebar and type in terminal) and entering the command:
python --version
If the version is 3.6 or above then you're all set. If nothing comes up then you need to install Python. If you've already installed Python and nothing is coming up then you likely need to add it to your system path.
If you instead see Python 2 then you can check if Python 3 is installed using:
python3 --version
If this is the case then we can alias pip and python to Python 3 in your .bashrc file or equivalent.
alias python='python3'
alias pip='python3 -m pip'
Open a terminal (Cmd+Space, type 'terminal', hit Return) and then run:
brew install python
Go to the Anaconda download page and download the installer for Python 3.7. Run the installer.
Go to the Python.org download page and download the latest installer. Run the installer.
You should already have git installed. If not, please ask for help!
We're going to need a few Python packages for today, they are:
ipython
numpy
matplotlib
jupyter
seaborn
pip is the default Python package manager. We'll be installing to the local user's Python environment in order to avoid any permissions issues.
pip install --user ipython matplotlib numpy jupyter
This is generally a good idea to avoid conflicts with any other package managers you might have.
If pip works but the packages do not appear when you invoke python then you may have a separate Python 2 and Python 3 installation. There are a few different fixes for this:
To directly invoke Python 3's pip you can:
pip3 install --user <packages>
Or use a particular python to invoke pip you can:
python -m pip install --user <packages>
Lastly you can alias the first command in your bashrc file (or equivalent depending on what shell you are using):
alias pip='python -m pip'
If you've used an anaconda or miniconda install you can also use the conda package manager. This shouldn't conflict with pip but sometimes has more recent releases (which may include essential bugfixes).
conda install ipython matplotlib numpy jupyter
Chocolatey is a package manager for Windows, using this will likely make the installation process much easier, but you can still do it manually if you want.
You can setup chocolatey using the instructions from (here)[https://chocolatey.org/install].
Obviously Python is important for a Python workshop, you can check if you already have Python by opening a terminal (Window + R then type in cmd.exe) and entering
python --version
If the version is 3.6 or above then you're all set. If nothing comes up then you need to install Python. If you've already installed Python and nothing is coming up then you likely need to add it to your system path.
If you have chocolatey then this is pretty easy.
choco install python
Once the installation has finished, close and re-open your terminal and see if Python is working.
If it's not then you might need to add it to your system path. This can be done using the SETX command.
SETX PATH "%PATH%;C:\<path to Anaconda>\scripts;C:\<path to anaconda>"
You will need to close and re-open your terminal for the changes to be loaded.
Anaconda is a Python package manager and environment manager all in one. If you have Chocolatey you can install it instead of the default Python with:
choco install anaconda3
If you want the minimalist version with fewer default packages, then you can use miniconda instead.
choco install miniconda3
Once the installation has finished, close and re-open your terminal and see if Python is working.
If it's not then you might need to add it to your system path. This can be done using the SETX command.
SETX PATH "%PATH%;C:\<path to Anaconda>\scripts;C:\<path to anaconda>"
You will need to close and re-open your terminal for the changes to be loaded.
You may want to later set up some proper conda environments, but for now we will simplify matters by using the default environment.
If you're not going down the Chocolatey route, you can instead pick install Python directly from (here)[https://www.python.org/downloads/]. When prompted, make sure that you add anaconda to your PATH environment variable.
This can be done using the SETX command.
SETX PATH "%PATH%;C:\<path to python>\scripts;C:\<path to python>"
You will need to close and re-open your terminal for the changes to be loaded.
And if you prefer the direct download of anaconda or miniconda you can get it from (here)[https://www.anaconda.com/download/]. You may need to manually add this to your PATH environment variable.
This can be done using the SETX command.
SETX PATH "%PATH%;C:\<path to Anaconda>\scripts;C:\<path to anaconda>"
You will need to close and re-open your terminal for the changes to be loaded.
You may want to later set up some proper conda environments, but for now we will simplify matters by using the default environment.
We're going to need a few Python packages for today, they are:
ipython
numpy
matplotlib
jupyter
seaborn
qiskit
pip is the default Python package manager. We'll be installing to the local user's Python environment in order to avoid any permissions issues.
pip install --user ipython matplotlib numpy jupyter
This is generally a good idea to avoid conflicts with any other package managers you might have.
If you've used an anaconda or miniconda install you can also use the conda package manager. This shouldn't conflict with pip but sometimes has more recent releases (which may include essential bugfixes).
conda install ipython matplotlib numpy jupyter
If you've having a particularly bad day you may need to build from source. This will almost never happen and is merely mentionned for completeness.
We're going to need a git client, the easiest way to do this is git bash for windows.
As before we can use chocolatey to install git for windows.
choco install git
Alternatively, you can manually install it from (here)[https://gitforwindows.org/].
We're going to need a few Python packages for today, they are:
ipython
numpy
matplotlib
seaborn
jupyter
qiskit
pip is the default Python package manager. We'll be installing to the local user's Python environment in order to avoid any permissions issues.
pip install --user ipython matplotlib numpy jupyter
This is generally a good idea to avoid conflicts with any other package managers you might have. If this is failing with pip not being detected then you've got another system path issue. One quick fix is to instead run:
python -m pip install --user <packages>
In the longer term you will want to add pip to your system path.
If you've used an anaconda or miniconda install you can also use the conda package manager. This shouldn't conflict with pip but sometimes has more recent releases (which may include essential bugfixes).
conda install ipython matplotlib numpy jupyter