4. The AiiDA python interface

AiiDA is written in python, and while the verdi command line interface provides handy shortcuts for everyday operations, AiiDA provides its full power only via the python interface.

4.1. Using the python interface

There are three practical ways of using the python interface:

  1. python scripts that import the aiida python package

  2. the interactive verdi shell

  3. Jupyter notebooks

For this tutorial we recommend using the verdi shell (explained below) but if you’re already familiar with Jupyter notebooks you may use those as well. We will get back to option 1 (python scripts) towards the end of the tutorial.

4.1.1. The verdi shell

The verdi shell is a customized ipython shell, where all the AiiDA classes, methods and functions are already accessible. Type in the terminal

$ verdi shell

The verdi shell is handy for everyday AiiDA-based operations, e.g. creating, querying and using AiiDA objects. You would typically use two terminals, one for the verdi shell and one to execute bash commands.

Note Press Ctrl+Shift+T in order to open a new terminal tab. Don’t forget to workon aiida in the new tab before using the shell.

4.1.2. Jupyter notebooks

Start a jupyter notebook server:

$ jupyter notebook

In the new browser window, select New -> Python 2 (top right corner). You are now inside a jupyter notebook, consisting of cells where you can type portions of python code. The code will not be executed until you press Shift+Enter from within a cell.

In order to load the same environment as in the verdi shell, type %aiida in the first cell and execute it.

Note The verdi shell and the jupyter notebook are completely equivalent. Use either according to your personal preference.

You will still need sometimes to type command-line instructions in the terminal. Either keep a terminal open on the side or use execute terminal commands directly from the verdi shell or jupyter notebook by prefixing the command by an exclamation mark:

!verdi profile list

4.2. Loading a node

Most AiiDA objects are represented by nodes, identified in the database by their PK number (an integer). You can access a node using the following command in the shell:

node = load_node(<PK>)

Note In the following, we’ll use a dark gray border to indicate python code and a light gray border to indicate the terminal.

Load one of the calculation nodes you played around with before. Then get the density computed with the command

node.out.output_parameters.get_dict()

You can also type

node.res.

and then press TAB to directly access the keys of the output dictionary.