By default, when you use pip to install packages, it will install them in a default location where all other packages are installed. This default behaviour may however not be desirable. The solution is to to create isolated virtual environments in which you can install packages. You can think of isolated environments as individual sandboxes. By having a separate virtual environment for each of your projects, you will keep the packages installed in each project separate from one another.

If you are using Python 2 or Python 3 (up to and including Python 3.2), install virtualenv using pip install virtualenv or pip3 install virtualenv, respectively. You will then be able to create and work with virtual environments via the virtualenv command.

If you are using Python 3.3 or a more recent version, support for creating virtual environments should already be installed and accessible via the python -m venv command (on some systems, the equivalent pyvenv command may also be available).

For the rest of the guide, python -m venv will be used as the command for creating virtual environments, with the assumption that you are using Python 3.3 or greater. However, for older Python versions, virtualenv should work the same.

The first step is to create a virtual environment which creates a new subdirectory at your specified location. It may be a good idea to have all your virtual environments under one sub-directory. Let's assume that you want the new virtual environment tabulate to be created in the subdirectory $HOME/work/virtualenvs where $HOME refers to the user's home directory on Linux and Mac OS X. The following command will carry out this operation:

$ python -m venv ~/work/virtualenvs/tabulate

If you are on Windows, assuming that you want the new virtualenv to be created under C:\work\virtualenvs, the equivalent command would be:

> python -m venv C:\work\virtualenvs\tabulate

A new virtual environment tabulate has now been created for us. To be able to install packages in it, you need to activate it.

On Linux/Mac OS X you can activate our virtual environment, tabulate via the following command:

$ source ~/work/virtualenvs/tabulate/bin/activate

On Windows, the command to activate the virtual environment is:

> \work\virtualenvs\tabulate\Scripts\activate

Now, you are in the virtual environment. To see the packages currently installed in the virtual enviroment, you can use the pip list command:

(tabulate) $ pip3 list
pip (7.1.0)
setuptools (18.0.1)

You can see that only two packages installed in the virtual environment. You can install our desired packages using pip now. As an example of a package, you will install and use the tabulate package. It is an easy way to display tabular data without having to do a lot of tedious work yourself.

(tabulate) $ pip install tabulate
Collecting tabulate
Using cached tabulate-0.7.5.tar.gz
Installing collected packages: tabulate
Running setup.py install for tabulate
Successfully installed tabulate-0.7.5

Our package is now installed, so try running a simple example to see how it works.

The program below will print a table consisting of the numbers in a list and how many times each appears as a table:

from collections import Counter
from tabulate import tabulate

def frequency_table(numbers):
    table = Counter(numbers)
    rows = []
    for number in table.most_common():
         rows.append([number[0], number[1]])
    print(tabulate(rows, headers=['Number', 'Frequency']))

if __name__=='__main__':
    scores = [7, 8, 9, 2, 10, 9, 9, 9, 9, 4, 5, 6, 1, 5, 6, 7, 8, 6, 1, 10]
    frequency_table(scores)

Save the above program to a file and run it while within the virtual environment. Run the python or python3 command and passing the file name you saved the program to as the first argument:

$ python <file.py>

You should see the following output when run:

Number    Frequency
--------  -----------
    9            5
    6            3
    1            2
    5            2
    7            2
    8            2
   10            2
    2            1
    4            1

Feel free to explore the tabulate package more to see the different types of tables you can create with it.

Once you are done working in the virtual environment, the deactivate command will deactivate the virtual environment:

(tabulate) $ deactivate

Over time, the number of virtual environments can increase and you may want to remove the ones you don't need anymore. To do so, delete the directory for the virtual enviroment. For example, for the tabulate virtual environment created above, delete the tabulate directory from $HOME/work/virtualenvs on Linux/Mac OS X or C:\work\virtualenvs on Windows.

To summarize, here are the steps when working with virtual environments:

• Create a virtual environment using python -m venv, pyvenv, or virtualenv
• Activate it using the activate script
• Work
• Deactivate using deactivate