CardioBERTa.es_enriched / training_data_stats.json
fernandogd's picture
Release CardioBERTa-based clinical concept disambiguation model
41f90d2 verified
Raw History Blame Contribute Delete
930 Bytes
{
"triplets": 83752,
"malformed_rows": 0,
"unique_triplets": 83752,
"duplicate_triplets": 0,
"unique_cuis": 83752,
"unique_anchors": 81757,
"unique_positives": 83570,
"unique_terms": 159688,
"unique_anchor_positive_pairs": 83664,
"ambiguous_terms_across_cuis": 2204,
"self_pairs_exact": 0,
"self_pairs_normalized": 5552,
"cuis_with_multiple_terms": 78200,
"terms_per_cui_mean": 1.9337090457541313,
"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": 6.056738943547617,
"positive_words_mean": 7.531652975451332,
"cuis_common_with_synonyms": 83752,
"cuis_added_vs_synonyms": 0,
"cuis_missing_vs_synonyms": 0,
"terms_common_with_synonyms": 159688,
"terms_added_vs_synonyms": 0,
"terms_missing_vs_synonyms": 0
}