Request for original dataset

[masashitsubaki]

I visited the web site of Alchemy but could not download the original dataset as described in this issue #17.

Actually, I have emailed to Dr. Guangyong (one of the paper's authors) and received that
``the order in CSV file is randomly permuted and the data (each molecular property) is normalized for the contest.''
So I would like to use the original data like the format of the QM9 dataset,
that is, the xyz coordinates, property values, and their units (e.g., eV and eV/atom) that are not normalized.

[peiguo-scut]

Thanks for your interests in our work, you can download our original dataset in the format of the QM9 from our official link https://alchemy.tencent.com/data/alchemy-v20191129.zip

Best Regards
Guangyong

[masashitsubaki]

Thank you very much for providing the dataset and now I could download it.
I have a question about the energy value in the Alchemy dataset.

In the QM9 dataset, almost all of the studies transform the internal energy (original provided value) into the atomization energy, learn the atomization energy, and evaluate its prediction performance (chemical accuracy is 1.0 kcal/mol) in their papers. This transformation is done by using the atomref information in the QM9 repository.
https://github.com/masashitsubaki/QuantumDeepField_molecule#datasets

Could you provide the atomref information used in the Alchemy? Or did the Alchemy contest use the internal energy directly instead of the atomization energy?