Token Classification
Transformers
Safetensors
PyTorch
Dutch
roberta
ner
pii
pii-detection
de-identification
privacy
healthcare
medical
clinical
phi
dutch
openmed
Eval Results (legacy)
Instructions to use OpenMed/OpenMed-PII-Dutch-FastClinical-Small-82M-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use OpenMed/OpenMed-PII-Dutch-FastClinical-Small-82M-v1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="OpenMed/OpenMed-PII-Dutch-FastClinical-Small-82M-v1")# Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("OpenMed/OpenMed-PII-Dutch-FastClinical-Small-82M-v1") model = AutoModelForTokenClassification.from_pretrained("OpenMed/OpenMed-PII-Dutch-FastClinical-Small-82M-v1") - Notebooks
- Google Colab
- Kaggle
| Classification Report for Dutch PII Detection | |
| Model: distilbert/distilroberta-base | |
| ============================================================ | |
| precision recall f1-score support | |
| BANKACCOUNT 0.50 0.38 0.43 136 | |
| BUILDINGNUMBER 0.67 0.75 0.71 111 | |
| CITY 0.73 0.89 0.80 334 | |
| CREDITCARD 0.45 0.64 0.53 86 | |
| DATEOFBIRTH 0.64 0.86 0.73 138 | |
| EMAIL 0.99 1.00 0.99 347 | |
| FIRSTNAME 0.74 0.87 0.80 460 | |
| LASTNAME 0.72 0.59 0.65 354 | |
| MASKEDNUMBER 0.91 0.90 0.90 77 | |
| PASSWORD 0.98 0.95 0.96 94 | |
| PHONE 0.97 0.99 0.98 230 | |
| SSN 0.87 1.00 0.93 338 | |
| STREET 0.67 0.50 0.57 123 | |
| USERNAME 0.98 0.95 0.96 367 | |
| ZIPCODE 0.87 0.93 0.90 193 | |
| micro avg 0.81 0.85 0.83 3388 | |
| macro avg 0.78 0.81 0.79 3388 | |
| weighted avg 0.81 0.85 0.83 3388 | |