Create and Manage Python virtual environment

In any fast paced AI startup, developers and data scientist have to manage a lot of repositories. For me personally, I manage 3 or sometimes 4 repositories at a time and switching between virtual environemts for these repositories is pain according to me, writing source deactivate to first deactivate the exisiting environment and then activating it by cding into the repo and source env/bin/activate.

To avoid this I wrote a function that will ease this pain.

For this to work, lets create a directory to manage and store all your virtual environments. I like to keep them separate.

mkdir ~/environments

Once this is done, paste the below code into your rc file. For linux users it will be .bashrc

create_venv(){
        # First deactivate any existing environment
        deactivate
        
        # check if folder exist
        if test -d ~/environments/"$1"; then
                echo "Environment already present, activating it"
                source ~/environments/"$1"/bin/activate
        else
                python3 -m venv ~/environments/"$1"
                echo "Activating environment"
                echo $1
                source ~/environments/"$1"/bin/activate 
        fi
}

Let me explain what the above script does. First it deactivates any existing environemt, second it checks if the argument to the create_env function exists that is, if you have a repo called example then create_env example will first check if example exist, if it does then it will activate the environment present for it in the ~/environment directory. If the environment is not present it will create a python environment (python3) for the above script and then activate it.

It is a very basic script.

Usage:

Go to any repository directory

create_env "repo/folder name"