Downloads · 30 days
34
0% of all-time downloads
French-TV-transcript-NER
Pclanglais/French-TV-transcript-NER
French-TV-transcript-NER is a token classification model from Pclanglais. Use it when you need labels on individual words, such as names. It is set up for transformers.
French-TV-transcript-NER is a named-entity recognition model trained specifically on French TV headlines and transcript.
All-time downloads
18K
Public
Parameters
278M
1.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.1 GB · 98%
At a glance
- Task
- Token Classification
- Library
- transformers
- Model type
- deberta-v2
- Access
- Public
- Created
- Jul 2, 2024
- Updated
- Jul 2, 2024
- SHA
- a1bceafd
Try a prompt
Model card
From the Hugging Face model README
French-TV-transcript-NER is a named-entity recognition model trained specifically on French TV headlines and transcript.
Given the format specificities, generalist multilingual or French model were unperforming. Additionally, the new model also provide additional set of entities useful in production (such as distinction between first name and last name).
Entities
The model covers twelve entities:
- First name (prenom)
- Last name (nom)
- Location (lieu)
- Country (pays)
- Organization (organisation)
- Event (evenement)
- Nationality (nationalite)
- Broadcast name (emission)
- Product (produit), such as technological production, medicine, etc.
- Law (loi)
- Cultural creation (creation), such as movie titles, novels, etc.
- Disease (maladie)
- Task
- Token Classification
- Library
- transformers
- Type
- deberta-v2
- Created
- Jul 2, 2024
- Updated
- Jul 2, 2024