Downloads · 30 days
6
0% of all-time downloads
en_pipeline
kriskyle/en_pipeline
en_pipeline is a token classification model from kriskyle. Use it when you need labels on individual words, such as names. It is set up for spacy. The card lists the license as cc-by-sa-4.0.
All-time downloads
1.6K
Public
Repo size
946 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
Other502 MB · 53%
At a glance
- Task
- Token Classification
- Library
- spacy
- License
- cc-by-sa-4.0
- Access
- Public
- Created
- Jan 19, 2023
- Updated
- Jan 19, 2023
- SHA
- ae6f5416
Try a prompt
Model card
From the Hugging Face model README
| Feature | Description |
|---|---|
| Name | en_pipeline |
| Version | 0.0.0 |
| spaCy | >=3.4.2,<3.5.0 |
| Default Pipeline | transformer, ner |
| Components | transformer, ner |
| Vectors | 0 keys, 0 unique vectors (0 dimensions) |
| Sources | n/a |
| License | n/a |
| Author | Kristopher Kyle, Hakyung Sung |
Label Scheme
<details>This model identifies and categorizes Argument Structure Constructions (ASCs). ASC types are marked on the main verb of the ASC.
<summary>View label scheme (9 labels for 1 components)</summary>| Component | Labels |
|---|---|
ner | ATTR, CAUS_MOT, DITRAN, INTRAN_MOT, INTRAN_RES, INTRAN_S, PASSIVE, TRAN_RES, TRAN_S |
Accuracy
| Type | Score |
|---|---|
ENTS_F | 91.65 |
ENTS_P | 91.53 |
ENTS_R | 91.78 |
TRANSFORMER_LOSS | 10943.24 |
NER_LOSS | 18950.33 |
- Task
- Token Classification
- Library
- spacy
- License
- cc-by-sa-4.0
- Languages
- en
- Created
- Jan 19, 2023
- Updated
- Jan 19, 2023