CardioBERTa.sv_enriched / training_data_stats.json
fernandogd's picture
Release CardioBERTa-based clinical concept disambiguation model
34c964b verified
Raw History Blame Contribute Delete
930 Bytes
{
"triplets": 80858,
"malformed_rows": 0,
"unique_triplets": 80858,
"duplicate_triplets": 0,
"unique_cuis": 80858,
"unique_anchors": 78767,
"unique_positives": 80488,
"unique_terms": 155374,
"unique_anchor_positive_pairs": 80723,
"ambiguous_terms_across_cuis": 2371,
"self_pairs_exact": 0,
"self_pairs_normalized": 3795,
"cuis_with_multiple_terms": 77063,
"terms_per_cui_mean": 1.9530658685596973,
"terms_per_cui_median": 2.0,
"terms_per_cui_p95": 2.0,
"terms_per_cui_max": 2,
"triplets_per_cui_mean": 1,
"triplets_per_cui_median": 1.0,
"triplets_per_cui_p95": 1.0,
"triplets_per_cui_max": 1,
"anchor_words_mean": 4.381743303074526,
"positive_words_mean": 5.122473966707067,
"cuis_common_with_synonyms": 80858,
"cuis_added_vs_synonyms": 0,
"cuis_missing_vs_synonyms": 0,
"terms_common_with_synonyms": 155374,
"terms_added_vs_synonyms": 0,
"terms_missing_vs_synonyms": 0
}