Dataset Viewer
Auto-converted to Parquet Duplicate
dataset
dict
links
dict
modalities
list
quality
dict
tables
list
{ "id": "qm7_reference_matrix", "title": "QM7 full original matrix reference", "description": "The original 7,165-record QM7 matrix dataset, including molecular geometry representations and atomization-energy labels. Suitable for reproducing matrix-based QM7 benchmarks.", "domain": "small_molecules", "categor...
{ "studies": [ { "title": "Rupp et al. Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning (2012)", "url": "https://doi.org/10.1103/PhysRevLett.108.058301" } ], "original_data": "https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/qm7.mat", "redistributor"...
[ { "kind": "file", "modality": "source files", "path": "files/qm7.mat", "url": "https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/qm7.mat", "description": "Original source file in its native format.", "bytes": 17884280 } ]
{ "grade": "unmeasured", "tags": [ { "key": "origin", "level": "info", "label": "computed", "detail": "Labels are computed (for example DFT or molecular dynamics)." } ] }
[]

QM7 full original matrix reference

The original 7,165-record QM7 matrix dataset, including molecular geometry representations and atomization-energy labels. Suitable for reproducing matrix-based QM7 benchmarks.

Part of the Molecular Datasets collection.

Get started

Original files are available in their native formats.

Where the data comes from

Files and modalities

Modality Where Content
source files files/qm7.mat (18 MB) Original source file in its native format.

Download an individual file:

from huggingface_hub import hf_hub_download
path = hf_hub_download("liuganghuggingface/moldata-qm7_reference_matrix", "files/qm7.mat", repo_type="dataset")

Data quality

  • computed: Labels are computed (for example DFT or molecular dynamics).

Dataset notes

  • The original reference has 7,165 records with Coulomb matrices, coordinates, and energies. The separate DeepChem SMILES table has 6,834 rows and uses original identifiers.

License

Dataset-wide license: unspecified. Original source.

Citation

Cite the original study and data source:

Downloads last month
105