Token Classification
GLiNER
PyTorch
English
nvidia
PII
PHI
GLiNER
information extraction
entity recognition
privacy
Instructions to use nvidia/gliner-PII with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- GLiNER
How to use nvidia/gliner-PII with GLiNER:
from gliner import GLiNER model = GLiNER.from_pretrained("nvidia/gliner-PII") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -122,11 +122,11 @@ Synthetic persona-grounded dataset generated with NVIDIA NeMo Data Designer, spa
|
|
| 122 |
**Evaluation Results** <br>
|
| 123 |
From the combined evaluation across Argilla, AI4Privacy, and Gretel PII datasets:
|
| 124 |
|
| 125 |
-
| Benchmark |
|
| 126 |
-
| --------------------| -----------
|
| 127 |
-
| Argilla PII |
|
| 128 |
-
| AI4Privacy |
|
| 129 |
-
| nvidia/Nemotron-PII |
|
| 130 |
---
|
| 131 |
We evaluated the model using `threshold=0.3`. <br>
|
| 132 |
|
|
|
|
| 122 |
**Evaluation Results** <br>
|
| 123 |
From the combined evaluation across Argilla, AI4Privacy, and Gretel PII datasets:
|
| 124 |
|
| 125 |
+
| Benchmark | Strict F1 |
|
| 126 |
+
| --------------------| -----------: |
|
| 127 |
+
| Argilla PII | 0.70 |
|
| 128 |
+
| AI4Privacy | 0.64 |
|
| 129 |
+
| nvidia/Nemotron-PII | 0.87 |
|
| 130 |
---
|
| 131 |
We evaluated the model using `threshold=0.3`. <br>
|
| 132 |
|