eval.py

import argparse, csv, json
def load_schema(p): return json.load(open(p,'r',encoding='utf-8'))
def tag_text(text, schema):
    text_l=text.lower(); preds=set()
    for tag,kws in schema.items():
        for kw in kws:
            if kw in text_l: preds.add(tag); break
    return preds
def main():
    ap=argparse.ArgumentParser()
    ap.add_argument('--notes', required=True)
    ap.add_argument('--schema', default='tag_schema.json')
    args=ap.parse_args()
    schema=load_schema(args.schema)
    rows=list(csv.DictReader(open(args.notes,encoding='utf-8')))
    tp=fp=fn=0
    for r in rows:
        gold=set([t.strip() for t in r['gold_tags'].split(';') if t.strip()])
        pred=tag_text(r['note'], schema)
        tp+=len(gold&pred); fp+=len(pred-gold); fn+=len(gold-pred)
    prec = tp/(tp+fp) if (tp+fp)>0 else 0.0
    rec  = tp/(tp+fn) if (tp+fn)>0 else 0.0
    f1   = 2*prec*rec/(prec+rec) if (prec+rec)>0 else 0.0
    print(f"Precision: {prec:.3f}\nRecall: {rec:.3f}\nF1: {f1:.3f}")
if __name__=='__main__': main()