beki's picture
Update README.md
20fb59f
---
tags:
- flair
- token-classification
- sequence-tagger-model
language: en
widget:
- text: >-
SELECT shipping FROM users WHERE shipping = '201 Thayer St Providence RI
02912'
license: mit
datasets:
- beki/privy
---
| Feature | Description |
| --- | --- |
| **Name** | `en_spacy_pii_distilbert` |
| **Version** | `0.0.0` |
| **spaCy** | `>=3.4.1,<3.5.0` |
| **Default Pipeline** | `transformer`, `ner` |
| **Components** | `transformer`, `ner` |
| **Vectors** | 0 keys, 0 unique vectors (0 dimensions) |
| **Sources** | Trained on a new [dataset for structured PII](https://huggingface.co/datasets/beki/privy) generated by [Privy](https://github.com/pixie-io/pixie/tree/main/src/datagen/pii/privy). For more details, see this [blog post](https://blog.px.dev/detect-pii/) |
| **License** | MIT |
| **Author** | [Benjamin Kilimnik](https://www.linkedin.com/in/benkilimnik/) |
---
## English PII in Flair
This is the large 5-class NER model for English trained on protocol trace data generated by [Privy](https://github.com/pixie-io/pixie/tree/main/src/datagen/pii/privy/)
F1-Score: **0.9522**
Predicts 5 tags:
| **tag** | **meaning** |
|---------------------------------|-----------|
| PER | person name |
| LOC | location name |
| ORG | organization name |
| DATE_TIME | dates and times |
| NRP | nationalities, religious and political groups |
Uses distilbert embeddings.
---