Dmitry43243242/icd10-ru-subgroup-c

ICD-10 subgroup classifier - group C (distilled specialist)

Multi-label classifier over 3-character ICD-10 subgroups inside chapter C. This specialist was distilled from local BERT teacher models into alexyalunin/RuBioBERT. Teacher weights are not uploaded to Hugging Face.

Intended use / Назначение

• EN: Decision-support signal for suggesting candidate ICD-10 subgroups from Russian clinical notes. Not a substitute for clinician judgment; not validated for autonomous diagnosis.
• RU: Вспомогательный сигнал для предложения кандидатных 3-символьных кодов МКБ-10 по русскому клиническому тексту. Не заменяет врача и не предназначен для автономных клинических решений.

Training data / Обучающие данные

• Source CSV: datasets/subgroups/group_C.csv
• SHA-256: 9526d7bd571f6aa94d0e162b727474a36dc63f71f79f0d78f400195b786bec26
• Splits: train=346 · val=75 · test=75
• Labels: 70; rare/interface-only ids are listed in label_map.json.

Training route

• Approach: direct_hard_training_no_distillation
• Base model: alexyalunin/RuBioBERT
• Direct validation hit@3: 0.96
• No-distillation threshold: 0.9
• Teacher models (fallback KD only): []
• Selected KD config (fallback only): temperature=None, hard_loss_weight=None

Metrics (test split)

metric	final specialist	teacher ensemble / fallback
macro_f1	0.5899
micro_f1	0.5926
weighted_f1	0.6300
subset_accuracy	0.2133
hit@1	0.8667
hit@3	0.9200
recall@3	0.8813
mrr	0.9023

Full per-label breakdown is available in metrics.json.

Inference

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

repo = "Dmitry43243242/icd10-ru-subgroup-c"
tok = AutoTokenizer.from_pretrained(repo)
mdl = AutoModelForSequenceClassification.from_pretrained(repo)
mdl.eval()

text = "жалобы пациента..."
inp = tok(text, return_tensors="pt", truncation=True, max_length=512)
with torch.no_grad():
    probs = torch.sigmoid(mdl(**inp).logits)[0]
preds = [mdl.config.id2label[i] for i, p in enumerate(probs.tolist()) if p >= 0.5]
top5 = sorted(
    [(mdl.config.id2label[i], p) for i, p in enumerate(probs.tolist())],
    key=lambda x: -x[1],
)[:5]
print(preds, top5)