Virtual Environments are helpful in keeping all the dependencies (packages), required for a Python project, in one place. So you can avoid issues with conflicting dependencies when working on different projects simultaneously. For instance, one major use case is when you want to run different Python versions like 2.7, 3.5, and 3.6 on the same system.

You can use the following steps when setting up any Python project (including web development project using frameworks like Django, Flask etc.)

1. Make a top-level Directory

Create a directory named “virtualenvs” (a parent directory to keep all your virtual environments). All your Python projects will live inside this directory.
mkdir virtualenvs

2. Create the Python Virtual Environment

cd virtualenvs
python3 -m venv env

You may use any name for the directory, other than “env”. If you’re curious about what goes inside the auto-generated “env” directory, you can check here:
cd env
cat pyvenv.cfg
python —version
pip —version

3. Activate the virtual env

On Mac: source bin/activate
On Linux: ./bin/activate
On Windows: .\Scripts\activate

Important: You need to activate the virtual environment every time you want to build or run the code. You may edit the code files without activating the venv.
Use deactivate to leave the virtual environment, or simply close the session/Terminal.

4. Create the project directory

mkdir your_proj_name

Optionally, Initialise git inside your_proj_name directory (you don’t need to commit the other files of env directory into git.):

git init
OR if you have the project code already, git clone the code inside env:
git clone [url]

5 a. Install packages/dependencies using pip

If you’ve a requirements.txt file:
pip install -r requirements.txt

or install the packages individually, for example for Django:
pip install django or
pip install django==2.1 # For a particular version

5 b. You can also install packages from Source Code

For example download Django pre-release code from their git repo, and unzip it into a directory named django, then:

pip install -e django
OR
pip install --editable django

Note that you may run pip from any of env‘s subdirectories: the installed code will always go into the env/lib and env/bin directories.

6. Finish and Run

You’re set to start working on the project now.
If you’re satisfied with the setup, you most probably want to keep a list of dependencies you just installed.

pip freeze > requirements.txt

(Keep the requirements file in git if you’ve initialised the git repo.)

7. Optimizing your daily workflow

I made this small shell script to automate the task of entering and activating the venv every-time I start:

#!/bin/bash
cd virtualenvs/env/
source bin/activate
# Configure any other environment variables here -
# for example, private API keys that you don't want to check in to git.
cd your_proj_name/

Paste the above code in a file name  start_your_project_name.sh. From now on, to get started, you just have to do this:

source start_your_project_name.sh
or ./start_your_project_name.sh