QM7b and Showcase: Dipole Polarizabilities for Small Molecules

Dipole polarizabilities for 7,211 molecules from the QM7b dataset and 52 molecules from a showcase dataset. The data was computed using linear response coupled cluster theory and density functional theory with a d-aug-cc-pVDZ basis set. The dataset was contributed by Yang Yang of Cornell University.

Use Cases

• Benchmarking quantum chemistry methods based on high-accuracy coupled-cluster polarizability calculations.
• Training machine learning models to predict polarizability based on molecular structure.
• Comparing density functional theory performance against coupled-cluster theory for molecular property prediction.
• Studying structure-property relationships for small organic molecules.

Strengths

• Includes polarizabilities for 7,211 molecules from the established QM7b benchmark.
• Provides high-accuracy reference values computed with coupled-cluster theory.
• Offers a direct comparison with density functional theory results for the same molecules.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the main dataset is known, but the total number of data points per molecule is unspecified.
• Last update date is unknown; freshness unverified.

Provenance

Source

Yang Yang, Cornell University, via paperswithcode.

Collection Method

Computational quantum chemistry calculations on existing molecular datasets.