Ingesting Data From QM9

QM9 is a computational quantum chemistry dataset of molecules, where each molecule is saved as an XMol file. A subclass of the pypif ChemicalSystem (XMolMolecularSystem) represents each entry and implements methods for loading System objects from XMol files. Doc strings document the system attributes and properties that are available for each entry. The molecular structure is made up of XMolAtomicSystem sub-systems, which have their own (documented) attributes and properties.

The dataset can be analyzed by loading the System entries into a PifFrame, which is a subclass of the pandas DataFrame object. One of the main advantages of the PifFrame is that is can be used to expose the sub-systems in a systems tree (i.e. each sub-system will get its own row instead of being imbeded in the row of its root system).

A subset of the QM9 dataset can be found in the data folder, and all base code and test stripts can be found in src.

The system_test.py script demonstrates:

1. How each molecular system can be loaded from files.
2. How PifFrame can be used to expose sub-systems.
3. How to write systems to JSON files in the PIF schema.