CardioBERTa.it_enriched / training_data_stats.json
fernandogd's picture
Release CardioBERTa-based clinical concept disambiguation model
cb019b0 verified
Raw History Blame Contribute Delete
925 Bytes
{
"triplets": 69631,
"malformed_rows": 0,
"unique_triplets": 69631,
"duplicate_triplets": 0,
"unique_cuis": 69631,
"unique_anchors": 68572,
"unique_positives": 69113,
"unique_terms": 136720,
"unique_anchor_positive_pairs": 69538,
"ambiguous_terms_across_cuis": 2244,
"self_pairs_exact": 0,
"self_pairs_normalized": 185,
"cuis_with_multiple_terms": 69446,
"terms_per_cui_mean": 1.9973431373957002,
"terms_per_cui_median": 2,
"terms_per_cui_p95": 2.0,
"terms_per_cui_max": 2,
"triplets_per_cui_mean": 1,
"triplets_per_cui_median": 1,
"triplets_per_cui_p95": 1.0,
"triplets_per_cui_max": 1,
"anchor_words_mean": 5.925406787206848,
"positive_words_mean": 6.174950812138272,
"cuis_common_with_synonyms": 69631,
"cuis_added_vs_synonyms": 0,
"cuis_missing_vs_synonyms": 0,
"terms_common_with_synonyms": 136720,
"terms_added_vs_synonyms": 0,
"terms_missing_vs_synonyms": 0
}