ner_evaluation.py

from spacy.scorer import Scorer

scorer = Scorer()

# Example evaluation data in the same format as training data
EVAL_DATA = TRAIN_DATA = [
    # GPS_COORDINATES
    ("Coordinates: 124.50 N, 68.95 W", {"entities": [(13, 29, "GPS_COORDINATES")]}),
    ("Location: 13.35 S, 57.80 E", {"entities": [(11, 24, "GPS_COORDINATES")]}),
    ("GPS: 24.50 N, 57.80 W", {"entities": [(5, 19, "GPS_COORDINATES")]}),

    # BANK_ACCOUNT
    ("Bank account: 1235-5778-1235-5778", {"entities": [(13, 33, "BANK_ACCOUNT")]}),
    ("Account number: 4332-8776-4332-8776", {"entities": [(16, 36, "BANK_ACCOUNT")]}),
    ("Acct: 1122-2233-3344-4455", {"entities": [(6, 26, "BANK_ACCOUNT")]}),

    # ID
    ("ID: ABC124", {"entities": [(4, 10, "ID")]}),
    ("Personal ID: XYZ790", {"entities": [(12, 18, "ID")]}),
    ("UID: JKL790", {"entities": [(5, 11, "ID")]}),

    # CREDIT_CARD
    ("Credit card: 1235 5779 9102 1122", {"entities": [(13, 33, "CREDIT_CARD")]}),
    ("Card details: 4332 8776 2102 1122", {"entities": [(14, 34, "CREDIT_CARD")]}),
    ("Card: 7891 1235 5679 9013", {"entities": [(6, 26, "CREDIT_CARD")]}),

    # ADDRESS
    ("Address: 124 Pine St.", {"entities": [(9, 22, "ADDRESS")]}),
    ("Home address: 457 Oak St.", {"entities": [(14, 27, "ADDRESS")]}),
    ("Office: 790 Maple St.", {"entities": [(8, 21, "ADDRESS")]}),
]



results = []

for text, annot in EVAL_DATA:
    doc_gold = nlp.make_doc(text)
    example = Example.from_dict(doc_gold, annot)
    doc_pred = nlp(text)
    result = scorer.score([example])
    results.append(result)

result