ipyoptimade

OPTIMADE Jupyter widgets and Voilà application

Query for and import structures from OPTIMADE providers (COD, MaterialsCloud, NoMaD, Materials Project, ODBX, OQMD, and more ...). The package provides a Jupyter widget for querying OPTIMADE providers and an example Voilà application to stack widgets into an web application.

Current supported OPTIMADE API versions: 1.1.0, 1.0.0, 1.0.0-rc.2, 1.0.0-rc.1, 0.10.1

Install with

pip install ipyoptimade

Run the client

This Jupyter-based app is intended to run either:

• In AiiDAlab as well as inside a Quantum Mobile Virtual Machine;
• As a MaterialsCloud tool;
• As a standalone MyBinder application; or
• As a standalone local application (see more information about this below).

For AiiDAlab, use the App Store in the Home App to install it.

Usage

AiiDAlab

To use the OPTIMADE structure importer in your own AiiDAlab application write the following:

from aiidalab_widget_base import OptimadeQueryWidget
from aiidalab_widgets_base.viewers import StructureDataViewer
from ipywidgets import dlink

structure_query = OptimadeQueryWidget()
structure_viewer = StructureDataViewer()

# Save to `_` in order to suppress output
_ = dlink((structure_query, 'structure'), (structure_viewer, 'structure'))

display(structure_query)
display(structure_viewer)

This will immediately display a query widget with a dropdown of current structure databases that implements the OPTIMADE API.

Then you can filter to find a family of structures according to elements, number of elements, chemical formula, and more. See the OPTIMADE API specification for the full list of filter options and their description.

In order to delve deeper into the details of a particular structure, you can also import and display OptimadeResultsWidget.
See the notebook optimade-client.ipynb for an example of how to set up a general purpose OPTIMADE importer.

Embedded

The query widget may also be embedded into another app.
For this a more "minimalistic" version of the widget can be used by passing embedded=True upon initiating the widget, i.e., structure_query = OptimadeQueryWidget(embedded=True).

Everything else works the same - so you would still have to link up the query widget to the rest of your app.

General Jupyter notebook

The package's widgets can be used in any general Jupyter notebook as well as AiiDAlab. Example:

from ipyoptimade import
    OptimadeQueryProviderWidget,
    OptimadeQueryFilterWidget,
    OptimadeSummaryWidget
from ipywidgets import dlink

database_selector = OptimadeQueryProviderWidget()
structure_query = OptimadeQueryFilterWidget()
structure_viewer = OptimadeSummaryWidget()

# Save to `_` in order to suppress output
_ = dlink((database_selector, 'database'), (structure_query, 'database'))
_ = dlink((structure_query, 'structure'), (structure_viewer, 'entity'))

display(database_selector, structure_query, structure_viewer)

This will use the package's own structure viewer and summary widget.

Note, the OptimadeQueryWidget mentioned above is a special wrapper widget in AiiDAlab for the OptimadeQueryProviderWidget and OptimadeQueryFilterWidget widgets.

Running application locally

To run the application locally, you need to have Jupyter installed. You can then run the application by opening the notebook optimade-client.ipynb in Jupyter and running all cells. If you have the voila package installed, you can also run the application in Voilà by clicking the Voilà button in the Jupyter notebook toolbar.

Configuration (Voilà)

For running the application (in Voilà) on Binder, the configuration file jupyter_config.json can be used.
If you wish to start the Voilà server locally with the same configuration, either copy the jupyter_config.json file to your Jupyter config directory, renaming it to voila.json or pass the configurations when you start the server using the CLI.

Note: jupyter_config.json is automatically copied over as voila.json when running the application using the optimade-client command.

Locate your Jupyter config directory:

jupyter --config-dir
/path/to/jupyter/config/dir

Example of passing configurations when you start the Voilà server using the CLI:

voila --enable_nbextensions=True --VoilaExecutePreprocessor.timeout=180 "OPTIMADE-Client.ipynb"

To see the full list of configurations you can call voila and pass --help-all.

Running with "development" providers (Materials Cloud-specific)

Set the environment variable ipyoptimade_DEVELOPMENT_MODE to 1 (the integer version for True (1) or False (0)) in order to force the use of development servers for providers (currently only relevant for Materials Cloud).

Development

Install with

pip install -e .[dev]
pre-commit install

Set

export ipyoptimade_DEBUG=1

to automatically open and show the debug & error messages in the OptimadeLog() widget.

Test voila with

voila optimade-client.ipynb

If dependencies are updated, update requirements.txt file that is used for the Docker image used in binder. The command to update the file is included at the top of requirements.txt.

Making a new release

To create a new release, clone the repository, install development dependencies with pip install -e '.[dev]', and then execute bumpver update [--major|--minor|--patch] [--tag-num --tag [alpha|beta|rc]]. This will:

1. Create a tagged release with bumped version and push it to the repository.
2. Trigger a GitHub actions workflow that creates a GitHub release and publishes it on PyPI.

Additional notes:

• Use the --dry option to preview the release change.
• The release tag (e.g. a/b/rc) is determined from the last release. Use the --tag option to switch the release tag.
• This package follows semantic versioning.

License

MIT. The terms of the license can be found in the LICENSE file.

Acknowledgements

	BIG-MAP; This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 957189. The project is part of BATTERY 2030+, the large-scale European research initiative for inventing the sustainable batteries of the future.

Contact

[e-mail]
[e-mail]