Token Classification
Transformers
Safetensors
PyTorch
German
deberta-v2
ner
pii
pii-detection
de-identification
privacy
healthcare
medical
clinical
phi
german
openmed
Eval Results (legacy)
Instructions to use kranushealth/OpenMed-PII-German-SuperClinical-Large-434M-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use kranushealth/OpenMed-PII-German-SuperClinical-Large-434M-v1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="kranushealth/OpenMed-PII-German-SuperClinical-Large-434M-v1")# Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("kranushealth/OpenMed-PII-German-SuperClinical-Large-434M-v1") model = AutoModelForTokenClassification.from_pretrained("kranushealth/OpenMed-PII-German-SuperClinical-Large-434M-v1", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| Classification Report for German PII Detection | |
| Model: microsoft/deberta-v3-large | |
| ============================================================ | |
| precision recall f1-score support | |
| ACCOUNTNAME 1.00 1.00 1.00 286 | |
| AGE 0.98 0.99 0.99 363 | |
| AMOUNT 0.99 0.99 0.99 127 | |
| BANKACCOUNT 1.00 1.00 1.00 303 | |
| BIC 1.00 1.00 1.00 76 | |
| BITCOINADDRESS 0.96 1.00 0.98 295 | |
| BUILDINGNUMBER 0.99 0.95 0.97 363 | |
| CITY 0.99 0.99 0.99 303 | |
| COUNTY 0.99 1.00 0.99 335 | |
| CREDITCARD 0.91 0.94 0.92 294 | |
| CREDITCARDISSUER 0.99 1.00 1.00 166 | |
| CURRENCY 0.77 0.89 0.83 213 | |
| CURRENCYCODE 0.99 1.00 0.99 75 | |
| CURRENCYNAME 0.52 0.35 0.42 78 | |
| CURRENCYSYMBOL 0.95 0.99 0.97 317 | |
| CVV 0.99 0.99 0.99 86 | |
| DATE 0.81 0.90 0.85 453 | |
| DATEOFBIRTH 0.82 0.75 0.78 348 | |
| EMAIL 1.00 1.00 1.00 449 | |
| ETHEREUMADDRESS 1.00 1.00 1.00 190 | |
| EYECOLOR 0.98 1.00 0.99 117 | |
| FIRSTNAME 0.99 0.98 0.99 1701 | |
| GENDER 1.00 1.00 1.00 312 | |
| GPSCOORDINATES 1.00 1.00 1.00 217 | |
| HEIGHT 1.00 1.00 1.00 110 | |
| IBAN 1.00 1.00 1.00 265 | |
| IMEI 1.00 1.00 1.00 249 | |
| IPADDRESS 1.00 1.00 1.00 779 | |
| JOBDEPARTMENT 0.99 1.00 0.99 330 | |
| JOBTITLE 1.00 1.00 1.00 333 | |
| LASTNAME 0.98 0.99 0.99 509 | |
| LITECOINADDRESS 1.00 0.85 0.92 75 | |
| MACADDRESS 1.00 1.00 1.00 123 | |
| MASKEDNUMBER 0.92 0.89 0.91 242 | |
| MIDDLENAME 0.94 0.98 0.96 330 | |
| OCCUPATION 0.99 1.00 0.99 344 | |
| ORDINALDIRECTION 1.00 1.00 1.00 141 | |
| ORGANIZATION 1.00 1.00 1.00 292 | |
| PASSWORD 1.00 1.00 1.00 303 | |
| PHONE 1.00 1.00 1.00 300 | |
| PIN 1.00 1.00 1.00 79 | |
| PREFIX 0.97 0.98 0.98 343 | |
| SECONDARYADDRESS 0.99 1.00 1.00 327 | |
| SEX 1.00 1.00 1.00 328 | |
| SSN 1.00 1.00 1.00 287 | |
| STATE 0.99 0.99 0.99 334 | |
| STREET 0.99 0.99 0.99 350 | |
| TIME 1.00 0.99 1.00 306 | |
| URL 1.00 1.00 1.00 288 | |
| USERAGENT 1.00 1.00 1.00 267 | |
| USERNAME 1.00 1.00 1.00 294 | |
| VIN 1.00 0.99 0.99 94 | |
| VRM 1.00 1.00 1.00 97 | |
| ZIPCODE 0.95 0.99 0.97 297 | |
| micro avg 0.97 0.98 0.98 15883 | |
| macro avg 0.97 0.97 0.97 15883 | |
| weighted avg 0.97 0.98 0.98 15883 | |