ORDerly supplementary datasets

Supplementary datasets used in ORDerly (i.e. the non-benchmark datasets)

• Condition prediction datasets: Contains parquet files for each of the four flavours of ORDerly-condition datasets that we used in the ORDerly paper.
• Condition prediction datasets config: Contains the .log and .json files showing the parameters used in cleaning and the impact on dataset size after each cleaning step.
• Transformer datasets: Contains plain txt files with the six transformer-ready datasets that were used for training/testing with Molecular Transformer.
• Non uspto data: Contains the datasets created with ORDerly from non-USPTO data in ORD. These datasets were used as test sets for forward prediction and retrosynthesis prediction.

Preprint: https://chemrxiv.org/engage/chemrxiv/article-details/64ca5d3e4a3f7d0c0d78ca42

Neurips workshop paper: https://openreview.net/forum?id=R8FQMsECIS

Code: https://github.com/sustainable-processes/orderly

The ORDerly benchmark datasets can be found here: https://figshare.com/articles/dataset/ORDerly_chemical_reactions_condition_benchmarks/23298467

Please feel free to contact me, Daniel Wigh, at dsw46@cam.ac.uk in case of any questions.