CardioBERTa.es_P_translations_only / training_data_stats.json
fernandogd's picture
Release CardioBERTa-based clinical concept disambiguation model
890337b verified
Raw
History Blame Contribute Delete
969 Bytes
{
"triplets": 1593029,
"malformed_rows": 0,
"unique_triplets": 1592694,
"duplicate_triplets": 335,
"unique_cuis": 476344,
"unique_anchors": 376578,
"unique_positives": 418463,
"unique_terms": 529722,
"unique_anchor_positive_pairs": 1576723,
"ambiguous_terms_across_cuis": 193752,
"self_pairs_exact": 0,
"self_pairs_normalized": 66,
"cuis_with_multiple_terms": 476343,
"terms_per_cui_mean": 3.9241199637236956,
"terms_per_cui_median": 3.0,
"terms_per_cui_p95": 9.0,
"terms_per_cui_max": 36,
"triplets_per_cui_mean": 3.34428270325647,
"triplets_per_cui_median": 2.0,
"triplets_per_cui_p95": 9.0,
"triplets_per_cui_max": 35,
"anchor_words_mean": 5.231707646251261,
"positive_words_mean": 5.421652085429707,
"cuis_common_with_synonyms": 69277,
"cuis_added_vs_synonyms": 407067,
"cuis_missing_vs_synonyms": 0,
"terms_common_with_synonyms": 136233,
"terms_added_vs_synonyms": 393489,
"terms_missing_vs_synonyms": 0
}