What you will learn in this chapter:
- how to set up a git repository,
- setup environment for development (I will use Ubuntu 16.04),
- install required packages,
- start the Django project.
Setup git repository
To set up a git repository I use GitHub (it is free for public and private projects). If you have an account there please go to https://github.com/new and set the repository, like in the image below.
The full code of this tutorial is available at: https://github.com/pplonski/my_ml_service.
Then please go to your terminal and set the repository:
git clone https://github.com/pplonski/my_ml_service.git cd my_ml_service ls -l
In my case, I had two files in the repository,
Let’s set up and activate the environment for development (I’m using Ubuntu 16.04). I will use
virtualenv venv --python=python3.6 source venv/bin/activate
You will need to activate the environment every time you are starting work on your project in the new terminal.
To install needed packages I will use
pip3 install django==2.2.4
The Django is installed in version
Start Django project
I will set up the Django project in the
backend directory. The Django project name is set to
mkdir backend cd backend django-admin startproject server
You can run your initiated server with the following command:
cd server python manage.py runserver
When you enter
127.0.0.1:8000 in your favorite web browser you should see default Django welcome site.
Congratulations!!! you have successfully set up the environment.
Add source files to the repository
Before we go to the next chapter, let’s commit new files.
# please execute it in your main project directory git add backend/ git commit -am "setup django project" git push
The following files should be added to your project:
new file: backend/server/manage.py new file: backend/server/server/__init__.py new file: backend/server/server/settings.py new file: backend/server/server/urls.py new file: backend/server/server/wsgi.py
In your directory, there are other files which are not added to the repository because there are excluded in
Next step: Train ML models