LSMO property visualizer app

Using

• bokeh server
• sqlite backend (AiiDA backend in alpha)

Configuration

The plots can be configured using yml files:

• figure/filters.yml: defines available filters
• figure/presets.yml: defines presets for axis + filter settings

Data

The data is availble on the Materials Cloud Archive. Place both the MOF_database.tar.gz and screening_data.tar.gz inside the data directory and run ./prepare.sh. This will unzip and build the database.

Installation

pip install -e .
./prepare.sh
./serve-app.sh
bokeh serve --show figure figure_top detail

Docker

The docker image should build successfully with; the database.db, and all cif files present in data/structures.

If this is not the case, the database can be built from the archive data.

Simply place the MOF_database.tar.gz and screening_data.tar.gz inside the data dir and run:

pip install -e . # Python3.7.15
./prepare.sh

If the correct data dir containing; database.db, 2 csvs and a large structures dir. The docker image can be built with all the data inside with

docker compose build
docker compose up
# open http://localhost:3247/select-figure

A built version of this docker image can be found on the private mc-Harbor docker registry.