CAMd Summer School 2024

Announcement: https://www.fysik.dtu.dk/english/research-sections/camd/events/summer-school-2024

Summer school exercises in Jupyter notebooks

The Summer School includes a number of Projects, which are partly formulated as Jupyter Notebooks. In a Jupyter Notebook you are running the calculations on the DTU central computing facilities, but the output is displayed in your local browser.

Unfortunately, this requires some setup, which is described below.

The computer system

The DTU computer system (known as the ‘G-databar’ for obscure reasons) consists of a login node (named login.gbar.dtu.dk or gbarlogin) and a number of compute nodes. Some of the compute nodes are reserved for batch jobs, some allow interactive jobs. You will be running a Jupyter Notebook server on an interactive compute node, this server will run the Python jobs and will allow the browser on your laptop to see the output. The latter unfortunately requires bypassing a firewall which would normally prevent you from accessing the compute nodes directly from the summer school site.

Instructions

The instructions depend on whether your laptop runs Windows, MacOS or Linux - the latter two are very similar in this context.

Windows users

• Setting up the first time (Windows)
• Starting and accessing a Jupyter Notebook (Windows)
• Submitting jobs on the DTU computers

Mac and Linux users

• Setting up the first time (Linux / macOS)
• Starting and accessing a Jupyter Notebook (Linux / macOS)
• Submitting jobs on the DTU computers

iPad users (not recommended)

• Using an iPad

Introductory slides

The slides from the introduction to the exercises (Sunday) can be found here: Intro_ASE_Databar.pdf

Projects

Choose a project according to your interests. The projects contain brief descriptions of what aspects of GPAW you learn from them.

• Projects
  • Introduction to Python and Jupyter notebooks
  • Catalysis
    • Part 1: N₂ adsorption on a flat Ru surface
    • Part 2: Splitting N₂: initial and final geometry
    • Part 3: Learning about Nudged Elastic Band
    • Part 4: Run a parallel NEB calculation
    • Extra exercise: Vibrational energy
    • Extra material: Convergence test
  • Magnetism in 2D
    • Part 1: Critical temperature of CrI₃
    • Part 2: Noncollinear magnetism in VI₂
    • Part 3: Find new ferromagnetic monolayers with high critical temperatures
  • Machine Learning
    • Part 1: Inspection of database
    • Part 2: Machine Learning
    • Part 3: Test and Evaluate the Model
  • Excited States
    • Part 1: Setup of the structure and bandstructure calculations
    • Part 2: Quasiparticle bandgap
    • Part 3: Absorption spectrum
    • Part 4: Excitonic effects and Discussion
  • Batteries
    • Part 1: Li intercalation energy in graphite
    • Part 2: Equilibrium potential of a LiFePO₄/C battery
    • Part 3: Transport barriers and voltage profile
  • Computational workflows
    • Part 0: The TaskBlaster tutorials and documentation
    • Part 1: Simple materials workflow
    • Part 2: Add ground state and band structure tasks
    • Part 3: Parametrize workflow over multiple materials

Instructions for instructors

Instructions for the people testing and developing the projects can be found here:

• Instructions for instructors
  • The DTU databar
    • Logging in
    • Creating an SSH key pair (while at DTU!)
      • Creating a key pair on Mac or Linux
      • Creating a key pair on Windows
    • Setting up the Virtual Environment etc
  • Summerschool Notebooks
    • Making small modifications to the notebooks
    • Teacher versus student version
    • Making new notebooks or large modifications
    • Running Notebooks in the Databar
    • Updating the text of the project pages