enze-chen / learn-citrination

Training materials

Repository from Github https://github.comenze-chen/learn-citrinationRepository from Github https://github.comenze-chen/learn-citrination

learn-citrination

Demos and tutorials for API access to Citrination. Documentation for the Python Citrination Client can be found here.

Contents

Four of the tutorials are organized as a sequence demonstrating the import and usage of DFT data:

  1. Importing VASP calculations
  2. Working with PIFs
  3. Introduction to queries
  4. Machine learning on Citrination

There are also advanced topics tutorials:

Requirements

Most of these tutorials are Jupyter notebooks backed by a python3 kernel. You'll need:

  • python3 with Jupyter. Anaconda is highly recommended.
  • Additional packages, which can be installed using pip:
pip install -U -r requirements.txt   

or conda:

while read requirement; do conda install --yes $requirement; done < requirements.txt
  • A valid Citrination Client API key set in your environment variables:

    1. Create an account on Citrination (if you don't already have one)
    2. Go to your account page and look for "API Key"
    3. Add the key to your environment. If you use a bash shell, the command is:

    export CITRINATION_API_KEY=<your key copied from the website>

    You can add this command to your .bashrc file to automatically set the API key in each new session.

    The API key functions as the password for the client, so it's important to keep it private. It's a good practice to put it in your environment instead of in the source code to help avoid unintentional sharing or publication of your API key.

Additional Resources

  • The example data used in these tutorials is drawn from: Alloy Database by Mihalkovic, Widom, and coworkers.

  • More API examples have also been developed by the Citrine Community team.

  • If you're looking for tutorials for the Citrination web user interface, please see this page.

About

Training materials


Languages

Language:Jupyter Notebook 61.4%Language:HTML 38.4%Language:Python 0.2%Language:Shell 0.0%