Data for canonical name translation experiments

• Full dump of ParaNames used to create parallel data for experiments

  • TSV formatted (full_paranames_dump.tsv.tar.gz)
  • DuckDB database (full_paranames_dump_duckdb.tar.gz, experimental)

• Parallel data used in experiments (parallel_data_for_experiments.tar.gz)

• Metadata (metadata.tar.gz)

  • Wikidata ID => train/dev/test split mapping
  • Sizes of each language in each split