Downloads · 30 days
127
0% of all-time downloads
en_spacy_pii_distilbert
beki/en_spacy_pii_distilbert
en_spacy_pii_distilbert is a token classification model from beki. Use it when you need labels on individual words, such as names. It is set up for spacy.
All-time downloads
96.6K
Public
Repo size
748 MB
Likes
35
Public
Repo makeup
Click a slice to open those files.
.whl486 MB · 65%
At a glance
- Task
- Token Classification
- Library
- spacy
- Model type
- distilbert
- Access
- Public
- Created
- Oct 14, 2022
- Updated
- Nov 12, 2024
- SHA
- 1c5cddb5
Try a prompt
Datasets
Model card
From the Hugging Face model README
| Feature | Description |
|---|---|
| Name | en_spacy_pii_distilbert |
| Version | 0.0.0 |
| spaCy | >=3.4.1,<=3.8.2 |
| Default Pipeline | transformer, ner |
| Components | transformer, ner |
| Vectors | 0 keys, 0 unique vectors (0 dimensions) |
| Sources | Trained on a new dataset for structured PII generated by Privy. For more details, see this blog post |
| License | MIT |
| Author | Benjamin Kilimnik |
Label Scheme
<details> <summary>View label scheme (5 labels for 1 components)</summary>| Component | Labels |
|---|---|
ner | DATE_TIME, LOC, NRP, ORG, PER |
Accuracy
| Type | Score |
|---|---|
ENTS_F | 95.42 |
ENTS_P | 95.30 |
ENTS_R | 95.54 |
TRANSFORMER_LOSS | 61154.85 |
NER_LOSS | 56001.88 |
- Task
- Token Classification
- Library
- spacy
- Type
- distilbert
- Languages
- en
- Created
- Oct 14, 2022
- Updated
- Nov 12, 2024