CardioBERTa.nl_enriched / training_data_stats.json
fernandogd's picture
Release CardioBERTa-based clinical concept disambiguation model
8e8b7d1 verified
Raw History Blame Contribute Delete
925 Bytes
{
"triplets": 79985,
"malformed_rows": 0,
"unique_triplets": 79985,
"duplicate_triplets": 0,
"unique_cuis": 79985,
"unique_anchors": 78049,
"unique_positives": 79788,
"unique_terms": 156019,
"unique_anchor_positive_pairs": 79931,
"ambiguous_terms_across_cuis": 2191,
"self_pairs_exact": 0,
"self_pairs_normalized": 1643,
"cuis_with_multiple_terms": 78342,
"terms_per_cui_mean": 1.979458648496593,
"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": 4.802537975870476,
"positive_words_mean": 5.751240857660812,
"cuis_common_with_synonyms": 79985,
"cuis_added_vs_synonyms": 0,
"cuis_missing_vs_synonyms": 0,
"terms_common_with_synonyms": 156019,
"terms_added_vs_synonyms": 0,
"terms_missing_vs_synonyms": 0
}